![Lexis, 14 | 2019, « Blending in English » [Online], Online Since 01 December 2019, Connection on 14 December 2020](http://data.docslib.org/img/a52c401e34debe461d2422fe23ac794b-1.webp)
Lexis Journal in English Lexicology
14 | 2019 Blending in English L'amalgamation en anglais
Isabel BALTEIRO and Laurie BAUER (dir.)
Electronic version URL: http://journals.openedition.org/lexis/1249 DOI: 10.4000/lexis.1249 ISSN: 1951-6215
Publisher Université Jean Moulin - Lyon 3
Electronic reference Isabel BALTEIRO and Laurie BAUER (dir.), Lexis, 14 | 2019, « Blending in English » [Online], Online since 01 December 2019, connection on 14 December 2020. URL : http://journals.openedition.org/lexis/ 1249 ; DOI : https://doi.org/10.4000/lexis.1249
This text was automatically generated on 14 December 2020.
Lexis is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. 1
The e-journal Lexis published its 14th issue, devoted to “Blending in English”, in December 2019. La revue électronique Lexis - revue de lexicologie anglaise a mis en ligne son numéro 14 en décembre 2019. Celui-ci est consacré à l'« amalgamation en anglais ».
Lexis, 14 | 2019 2
TABLE OF CONTENTS
Introduction Isabel Balteiro and Laurie Bauer
Papers
Blending creativity and productivity: on the issue of delimiting the boundaries of blends as a type of word formation Natalia Beliaeva
Improving on observational blends research: regression modeling in the study of experimentally-elicited blends Stefanie Wulff and Stefan Th. Gries
A corpus-based analysis of new English blends Mattiello Elisa
Variable base-word positioning in English blends Aviv Schoenfeld, Evan Gary Cohen and Outi Bat‑El
A study on the ‘wordgasm’: the nature of blends’ splinters Alejandro Barrena Jurado
Headedness in contemporary English slang blends Gorica Tomić
New lexical blends in The Simpsons: a formal analysis of English nonce formations and their French translations Adam Renwick and Vincent Renner
To blend so as to brand: a study of trademarks and brand names Jelena Danilović Jeremić and Jelena Josijević
Gold Punning: studying multistable meaning structures using a systematically collected set of lexical blends Daniel Kjellander
List of references Isabel Balteiro and Laurie Bauer
Lexis, 14 | 2019 3
Introduction
Isabel Balteiro and Laurie Bauer
1 Lexis 14 is devoted to the phenomenon of blending in English. Different motivations justify this choice: mainly, and despite the prolific literature on the topic, there are issues that still remain unsolved, but also the productivity of the mechanism as well as the growing number of lexical items created by this process that appear in daily in the world languages. This emergence and increasing appearance of blends is the reason for the synchronic approach adopted in the present volume as well as for its special focus on very recent study cases.
2 A lexical blend is generally defined as a word which cannot be analysed into morphemes (Bauer [1983: 234]; Cannon [1986]), intentionally formed by merging together elements or splinters usually from two source lexical units (sometimes more, e.g. afflufemza < affluence + influenza + feminism (example from Bassac [2012: 169]), or the more recent scinfotainment < science + information + entertainment). However, despite the recent interest in blending, and as already suggested, it is still a somehow poorly understood and underresearched mechanism, often regarded as “irregular” (Connolly [2013: 3]) and/or “marginal”. For these and other reasons, Lexis 14 aims at exploring the linguistic and even extralinguistic contexts which affect and motivate the creation and success of blends in English.
3 The 14th issue of Lexis focuses on the following three main areas of research: • The first area of research is centered on what constitutes a blend and the difficulties in distinguishing blends from other mechanisms, including problems posed by fuzzy boundaries, such as: the identification and limits of blending as either word creativity or word-formation; the identification of blending as a lexical or a semantic phenomenon or mechanism; or the differences between blending and clipped compounding (complex clipping). • The second area of research concentrates on the different types of blends: amongst other types, coordinate blends; headed blends; blends with truncation at the inner edge of the constituent words; or blends which do not contain any input word in its entirety. • The third area of research tackles the purposes and motivations for blending in English. Questions such as whether blends are created with the aim of designating a new referent or a new concept or to give a new name to an existing referent or concept; the need for an
Lexis, 14 | 2019 4
explanation of the varied motivations in different contexts and registers for the creations of blends; or how to measure the measure the weight of the semantic, phonological, graphemic and/or formal motivation of blending in English and compare and contrast the importance of meaning versus shape at the time of creating a blend.
4 These three areas of research account for the difficulties in delimiting and clearly defining blending and blends: first, the definition depends, on the one hand, on the fuzzy boundaries between word formation and word creativity and, on the other, on the existing controversies and fuzzy limits between blending and compounding, acronymy, etc. Apparently, there is not yet a clear answer to these questions, as lexicologists and morphologists vary in their answers and there is not general agreement as regards this issue. As a consequence of this, the different and emerging types of blends (cf., for example, Balteiro [2018]) are still to be explored and analyzed. Futhermore, the reasons and motivations behind the emergence, creation and rapid expansion of blending not only in different fields within a language but also in the world languages are also varied and contribute to the difficulties in analysing, defining and delimiting the concept. Blends seem to be everywhere, from the most technical language to the most informal, even in even slang, as will be seen in the following pages. Moreover, some are only created at a given moment and for a very specific purpose, almost at the spur of the moment, and, in this sense, are very close to neologisms (see Lexis 12) but probably with an even higher degree of ephemerality, while others are born to stay in the language.
5 The papers included in this issue (Lexis 14) are primarily based on the study of English blends; they are organized from the most general, dealing with the definition and delimitation of this type of word formation, then addressing questions such as the principles underlying this word formation or word creativity mechanism and its nature, to end with more specific issues and contexts where blends are highly productive such as informal and slang contexts, TV series, but also brand naming.
6 The first paper addresses and goes beyond the issue of delimiting the boundaries of blends as a type of word formation. Natalia Beliaeva, basing her arguments of both qualitative and quantitative analyses in “Blending creativity and productivity: on the issue of delimiting the boundaries of blends as a type of word formation”, also focuses on blending creativity, productivity and predictability factors that influence blend formation.
7 Stefanie Wulff & Stefan Th. Gries in the second paper of the volume, “Improving on observational blends research: regression modeling in the study of experimentally- elicited blends”, discuss the results of a blend production experiment and how it relates to previous research that was nearly exclusively based on observational data. They corroborate most of the following principles: (i) the shorter source word contributes more of itself to the blend than the longer source word, (ii) source word2 determines blend stress (more than source word1), and (iii) blending maximizes similarity between source words and blends.
8 The third paper of the volume, authored by Elisa Mattiello, “A corpus-based analysis of new English blends”, investigates new ‘attributive’ or ‘headed’ blends in English. Mattiello reassesses the importance of blending in terms of 1) its suitability in the coinage of new specialized vocabulary, and 2) its regularity in the creation of words containing frequent splinters. Furthermore, she also contributes to the issue of
Lexis, 14 | 2019 5
whether blending should be considered an extra-grammatical phenomenon of word- creation or a regular process of word-formation.
9 In “Variable base-word positioning in English blends”, the fourth paper of the volume, Aviv Schoenfeld, Evan Gary Cohen & Outi Bat El explore blend doublets and the conditions and factors that interact to allow variable positions of the blends’ base words.
10 Alejandro Barrena Jurado in his paper entitled “A study on the ‘wordgasm’: the nature of blends’ splinters”, from the theory of Construction Morphology, studies the ending –gasm to decide whether it is a splinter or a combining form and arrives at the conclusion that undoubtedly –gasm creates blends of varied types, which are then analysed.
11 Gorica Tomić in “Headedness in contemporary English slang blends”, the sixth article of the volume, addresses, as the title indicates, the issue of headedness. After carrying out both a qualitative and quantitative analysis, she concludes, amongst other ideas, that the semantic right-headedness is not as prominent as expected in slang blends.
12 Adam Renwick & Vincent Renner in “New lexical blends in The Simpsons: a formal analysis of English nonce formations and their French translations”, concentrate on the conspicuous presence of blends in The Simpsons to conclude that nonce blends are preferred to lexical shortening and that segment overlap and phonological headedness prevail in both English and French. Furthermore, against a widely-accepted principle among morphologists, they argue that blends may not constitute a homogeneous class from a formal standpoint.
13 The eighth article of the volume, “To blend so as to brand: a study of trademarks and brand names”, written by Jelena Danilović Jeremić & Jelena Josijević, explores the phonological, graphological, stylistic ad semantic motivations behind the creation of brand names through blending mechanisms.
14 Finally, in the last article of the volume, “Gold Punning: studying multistable meaning structures using a systematically collected set of lexical blends”, Daniel Kjellander aims at developing systematic approaches to the collection of lexical blends, as a necessary step in the development of blend research.
AUTHORS
ISABEL BALTEIRO Universidad de Alicante, Spain [email protected]
LAURIE BAUER Victoria University of Wellington, New Zealand [email protected]
Lexis, 14 | 2019 6
Papers Articles
Lexis, 14 | 2019 7
Blending creativity and productivity: on the issue of delimiting the boundaries of blends as a type of word formation
Natalia Beliaeva
Introduction
1 Setting the limits of blending as a type of word-formation has been a widely discussed question in morphological studies to date. In particular, whether blending is primarily a phenomenon of word creativity, or whether it is a regular and predictable mechanism of word formation, remains an open question. Rather than maintaining an argument in favour of either position, this study aims to investigate factors underlining such distinction, and to develop criteria that could be applied to corpus data to characterize particular examples of blending as points on a cline from word creativity to productive word formation.
2 As has been pointed out by a number of scholars (Algeo [1977], Dressler [2000], Fandrych [2008], Gries [2004], López Rúa [2004], Bauer [2012], to name just a few), blends are remarkably diverse in terms of their formal structure. While the majority of blends are formed out of two constituents by combining the initial segment of one with the final segment of the other, e.g. dramedy < drama + comedy, hangry < hungry + angry, there are a lot of counterexamples including formations that a) contain three or more constituents, e.g. Thankshallowistmas < Thanksgiving + Halloween + Christmas; b) involve intercalation of constituents, e.g. prowebstinate < procrastinate + web; c) combine final parts of both words, e.g. frohawk < afro + Mohawk; or d) demonstrate other unconventional ways of combining lexical items, e.g. dublexia < www + dyslexia. In addition, the constituents of blends display a variety of semantic relationships (see, for example, Algeo [1977] for classification). On the other hand, there is abundant evidence in the literature showing that the form of a blend can to a large extent be predicted
Lexis, 14 | 2019 8
based on phonological and phonotactic properties of their source words (Kubozono [1990], Gries [2012], Arndt-Lappe & Plag [2013], Beliaeva [2014]) and the similarity between them (Kelly [1998], Gries [2006, 2012]). The results of corpus studies such as those in Gries [2006, 2012], as well as experimental data such as those in Arndt-Lappe & Plag [2013] and in Beliaeva [2016] demonstrate that the formation of blends is subject to psycholinguistically relevant constraints such as recognisability of the source words.
3 In some cases, the formation of blends is not only predictable, but may give rise to productive constructions where part of a word that was once blended is used to form further blends, e.g. ‑cation (from vacation) in workcation, spa-cation, staycation, etc. As discussed in Lehrer [1998] and as further substantiated by contemporary corpus data in Bauer, Beliaeva & Tarasova [2019], the morphological status of formations containing splinters such as ‑cation can be compared to that of neoclassical compounds or, to some extent, affixations. At the same time, a lot of blends do not demonstrate such productivity, and factors that influence this are yet to be investigated. An aspect of the use of blends that is essential to consider is the playful character of many blends and their use as expressive means of language (see, for example, Lehrer [2007] and Renner [2015] for the discussion of blends as creative lexical formations). Importantly, factors that enhance the creative and attention-catching properties of blends may at the same time decrease the predictability of their form, as specified by Renner [2015].
4 This study argues that blending, as a word formation process, can be driven by either factors enhancing predictability of the outcome or factors enhancing playful character of the formation, or both. Corpus-based approach is used to investigate the relationship between well-formedness of the blends (that is, conforming to prosodic and recognisability constraints), and their productivity potential. Data from the Corpus of Contemporary American English (Davies [2008–]), earlier publications on blends, online collections of neologisms such as Word Spy, Urban Dictionary and The Rice University Neologisms Database, and other media sources are used to investigate exemplars of different structures of blends. Data from the NOW Corpus (Davies [2013–]) are further analyzed to construe patterns of recurrent use of blend constituents in various novel formations. The results are used to develop criteria of characterizing individual blends in terms of their productivity and creativity, and to substantiate the identification of blending as a mechanism involving both word creativity and productive word formation at varied degrees.
1. Fuzzy boundaries of blends revisited
5 A blend word used in advertisement, e.g. Frappuccino® (Urban Dictionary), or in political media, e.g. Brexit (Top Words for the first 15 Years of the 21st century and what they portend) is both attention-catching and thought-provoking. Most blends are attention-catching for language users, and the reasons for this will be considered in Section 2. For linguists, however, this phenomenon has been mind-bogging for decades, because the unusual formal properties of blend words made it difficult to provide an exhaustive description of blends as a word formation category or even define what a blend is. In this section, the much-discussed question of defining the status of blends among other types of word formation will be revisited not with the purpose of putting an end to the discussion, but, rather, to accentuate factors that are essential for understanding the connection between blends and other word formation types.
Lexis, 14 | 2019 9
6 Putting together two words to form a compound such as sugar bowl or blackboard can be considered one of the most straightforward ways to form a new lexeme – the universality of compounds is pointed out, for example, in Fromkin, Rodman & Hyams [2012]. A less straightforward and less common way of making a new lexeme using two existing ones is merging two (or sometimes more) words together so that part of the material is lost in the process, which will produce a blend. In fact, in earlier morphological studies such as those by Marchand [1969] and Adams [1973], blends have been classified as a subtype of compounds. Other scholars, such as Devereux [1984] and Cannon [1986], considered truncation a more important prerequisite of blending and therefore classified blending as a type of shortening. Since both shortening and concatenating are essential characteristics of blending, as admitted by many scholars, regarding blending as a word formation type akin to both compounding and shortening should not be a contradiction. For example, López Rúa [2004] categorizes blends as points on a cline between compounds and abbreviations, depending on how much the source words are truncated while forming a particular blend. Similarly, prototypical approach to classification of blends is undertaken by Brdar-Szabó & Brdar [2008] and by Mattiello [2013].
7 Even such a brief glimpse of the vast literature on blending would lead us to a conclusion that the boundaries of the category are fuzzy. What is more remarkable, some researchers claim that the formal diversity that precipitates the fuzzy boundaries is an essential characteristic of blending, while others focus on the regularities that can be observed in linguistic data, and aim to find grounds for a systematic description of blends drawing on those features of blends that are predictable, at least to some extent. The former approach is taken, for example, by López Rúa [2004, 2012] and by Mattiello [2013] whose classifications of blends are based on the results of analyzes involving a plethora of structurally different examples. Taken to extremes, the account of formal diversity of blends can lead to conclusions that they are an “extragrammatical” phenomenon (Dressler [2000]), which is, at best, “minimally predictable” (Mattiello [2013: 96]), and therefore blends can be regarded as instances of word-creation as defined by Ronneberger-Sibold [2010].
8 A factor that is often related to the diversity and unpredictability of blends is the punning nature of many blends if not all of them. Thus, Renner [2015: 130] claims that “[c]oining a new lexical blend is an act of wordplay”, having defined wordplay in the sense that bears some similarity with the concept of word-creation by Ronneberger- Sibold [2010], that is, as follows: […] an intentional and formally ingenious way of associating the semantics of two or more words in a new morphological object. [Renner 2015: 119]
9 Renner substantiates this claim by analyzing examples of lexical blends that involve various degrees of overlapping between the source words, infixation, intercalation, violations of structural well-formedness constraints that are otherwise observable in the language, graphic and semantic play on words. The importance to take into consideration the playful character of blends is reiterated in Beliaeva [2019], where it is also related to the use of blends as expressive means in various domains, e.g. in youth slang (01), journalism (02), political media (03) or as means of creating brand names (04) or other attention-catching titles such as musical band names (05) – note that parts of the words that are omitted in the process of blending are put in parentheses:
Lexis, 14 | 2019 10
(01) brovember < bro + (N)ovember, referring to a month of doing manly things (Urban Dictionary) (02) Brangelina < Br(ad) [Pitt] + Angelina [Jolie] (Urban Dictionary) (03) Merkozy < Mer(kel) + (Sar)kozy, referring to a unified position of France and Germany (Stewart [2011]) (04) Teaffee < tea + (co)ffee, the name of a café (http://teaffee.com/ index.html) (05) Stratovarius < Strato(caster) + Stra(di)varius, the name of a musical band combining heavy metal and classical music (López Rúa [2012: 31])
10 As further observed in Beliaeva [2019: n.p.], wordplay cannot be regarded as “the only driving force of blend formation”, because in such cases as creating names of animal hybrids (06), language varieties (07) or hybrid recipes (08) lexical blending may be used in order to reflect the hybrid nature of the objects that are named.
(06) zorse < z(ebra) + (h)orse, referring to a hybrid of a zebra and a horse (Urban Dictionary) (07) Spanglish < Span(ish) +(En)glish, used to refer to dialects, pidgins, or creole languages that result from interaction between Spanish and English (Nash [1970]) (09) cronut < cro(issant) + (dough)nut, a doughnut-croissant hybrid (Merwin [2013])
11 Although language play thus cannot be seen as the only factor that influences the formation of blends, it should definitely be taken into consideration in the analysis of blending as a morphological phenomenon.
12 Scholars arguing for predictability of blends focus on the regularities that may help develop a systematized account of this category, despite its fuzzy boundaries. Such systematizations have been developed in many studies including Kubozono [1990], Gries [2006, 2012], Arndt-Lappe & Plag [2013] and Beliaeva [2014]. Thus, in Gries [2006, 2012] and Beliaeva [2014], inferences about systematic nature of blends are made based on corpus data on the frequency of occurrence of certain types of formations. Some insights into the mechanism of blending are drawn from considering cognitive factors involved in the formation (Arndt-Lappe & Plag [2013], Beliaeva [2014], Gries [2006, 2012]) and processing (Beliaeva [2015, 2016]) of blends.
13 In sum, two important directions of reasoning can be distinguished in the abundant literature on blends. On the one hand, investigating formal attributes of blends to the full extent of their diversity can provide valuable information about the genesis of blends and their role in language. On the other hand, insights from frequency-based accounts of regularities of blends are no less valuable. However, it appears that focussing on one of these two aspects of the phenomenon may lead to blinkered vision of the complex picture. In Sections 2 and 3, we will consider both the diversity and the predictability of blends, in an attempt to provide a plausible explanation of the fuzzy boundaries of the category and uncover some factors involved in blend formation that may have been understudied to date.
Lexis, 14 | 2019 11
2. Multiformity of blends as interplay between creativity and predictability
2.1. Formal diversity of blends
14 In most general terms, the mechanism of blending was described in Plag [2003: 123]: AB + CD = AD
15 Thus blending involves combining parts of two source words so that the beginning of the first one is conjoined with the ending of the second one. However, the linguistic examples that appear to accord with this seemingly simple formula exhibit exceptional formal diversity. As summarized in Beliaeva [2019], following earlier classifications in Algeo [1977], Cannon [1986], Gries [2006] and other studies, the possible outcomes of merging a word AB with a word CD include cases when the beginning of one source words is concatenated with the ending of another one (10), when the same process takes place but the source words overlap where they are merged together (11), when the source words overlap so that the first word is entirely preserved in the blend (12), when the source words overlap so that the second word is entirely preserved in the blend (13), and when the source words overlap so that both of them are preserved in their entirety (14). The overlapping segments in the examples henceforth are in bold type:
(10) tigon < tig(er) + (li)on (COCA) (11) motel < mot(or) + (h)otel (COCA) (12) mockbuster < mock + (bl)ockbuster (Urban Dictionary) (13) jumbrella < jumb(o) + umbrella (Word Spy) (14) alcoholiday < alcohol + holiday (Urban Dictionary)
16 This classification is far from exhaustive, as less frequently attested outputs of merging a word AB with a word CD include cases when one word replaces a segment in the middle of another word (15), referred to as infixed blends in Bauer [2012] or as central replacement blends in Beliaeva [2014]. Sometimes the insertion takes place where the two words are phonetically and graphically similar, which thus involves some overlap (16). More rare cases include formations combining final parts of the source words (17– 18).
(15) pro-web-stinate < pro(cra)stinate + web (The Rice University Neologisms Database) (16) li-boob-rian < lib(ra)rian + boob (The Rice University Neologisms Database) (17) frohawk < (a)fro + (M)ohawk (Urban Dictionary) (18) podestrian < (i)Pod + pedestrian (Urban Dictionary)
17 Occasionally, more than two words are merged into one, such attestations ranging from blends with three elements (19) to more extreme cases of blending six words (20).
(19) Japornimation < Jap(a)n + porn + (a)nimation (Mattiello [2013: 122]) (20) Christmahanukwanzadandiwalstice < Christma(s) + Hanuk(kah) + Kwanza(a) + (Ram)adan + Diwal(i) + (Winter sol)stice (Renner [2015: 126])
18 Although deviating from a more widely attested pattern of coining blends by merging two words, formations such as those in (17–20) bear similarity with more typical blends in other respects. Thus, the source words of blends in (17–20) are merged together in
Lexis, 14 | 2019 12
points where the degree of similarity between them is maximal, and blends are coined in such a way that the source words remain recognisable in the resulting formations. It is worth noting that similarity and recognisability of the source words are factors that have been characterized by some scholars, such as Gries [2006, 2012] and Beliaeva [2014, 2016], as essential properties of typical blends that distinguish them from other types of word formation including clippings. With regard to the degree of similarity between the source words and the relative amount of the source word material preserved in the blend, formations in (17–20) can be considered more conforming to typical properties of blends than coinages combining initial parts of two (21) or more (22–23) source words.
(21) globfrag < glob(alisation) + frag(mentation) (Urban Dictionary) (22) SoLoMo < so(cial) + lo(cal) + mo(bile) (Word Spy) (23) COSPAR < Co(mmittee) (on) Spa(ce) R(esearch) (López Rúa [2002: 46])
19 Formations that are structurally similar to the example in (21) have been described in literature as clipping compounds (Bauer [2012]) or complex clippings (Gries [2006]). The formation in (22) is also similar to (21) but combines initial segments of three words rather than two. This example is also formally similar to acronyms as described, for example, in Quirk, Greenbaum, Leech & Svartvik [1985] and in López Rúa [2002] – also compare (22) to the formation in (23) which was provided as an example of an acronym. An analysis of comparable cases is undertaken in López Rúa [2004], where a prototypical account of blends, compounds and various types of shortenings is used to construct a “categorial continuum” of forms coined by concatenation or merging together parts of existing words. This categorial continuum arranges compounds, neo- classical compounds, blends and acronyms depending primarily on the degree of shortening, that is, on a cline from full preservation of constituents (the case of compounds) to minimal preservation (the case of acronyms preserving one letter from each of their source words such as NATO). A similar conclusion is drawn in Bauer [1998], on the basis of a study of neoclassical compounds. In particular, Bauer [1998: 414] considers blends and clipping compounds as “intermediate stages” between compounds and abbreviations in terms of the degree of shortening involved in their formation. A visualization of such account of blends and related morphological categories is proposed in Figure 1.
Lexis, 14 | 2019 13
Figure 1. Different structural types of blends in relation to additive and subtractive word formation categories
20 As is illustrated in Figure 1, the result of concatenating parts of several existing words can resemble a compound or an acronym to a certain extent, depending on how much material of the source words is preserved. The central area in Figure 1 represents cases that are often referred to as prototypical blends, that is, formations combining the beginning of one word with the ending of another, with some overlap in the middle. Peripheral cases include formations that are similar to acronyms, those similar to compounds, and those which combine word parts in a way that deviates from the more frequently observed pattern exemplified in the middle part of the figure. The more frequently observed pattern, in its turn, is subject to several factors that will be further discussed in Section 2.2.
2.2. Well-formedness and creativity of blends: A seeming paradox
21 Criteria of well-formedness or prototypicality of blends have been investigated in many studies (Marchand [1969], Kubozono [1990], Kelly [1998], Adams [2001], López Rúa [2004], Gries [2006, 2012], Bauer [2012], to name a few). These criteria of prototypicality concern semantic relations between the source words of the blend, as well as phonotactic and prosodic restrictions the blends are subject to. The semantic considerations of blend formation are beyond the scope of the present study (but see Devereux [1984], Cannon [1986], Renner [2006] for semantic classifications, Gries [2012] and Beliaeva [2014] for relevant quantitative analysis, and Bauer [2012] for a comprehensive summary).
22 The phonotactic and prosodic factors that are related to well-formedness of blends determine what components of the source words (such as syllable elements and main stress) are preserved in the blend. In particular, blends tend to preserve as much material of both of their source words as possible, while conforming to overall phonotactic constraints of language. Thus, typical blends (as discussed in more detail in Beliaeva [2019], following earlier investigations by Kubozono [1990], Kelly [1998], Arndt-Lappe & Plag [2013] and other scholars) have the same number of syllables as the
Lexis, 14 | 2019 14
longest of their source words (24–25, 27), or are one syllable longer (26). As observed in Beliaeva [2019], in many cases longer blends are formed due to preservation of maximal possible overlapping segment, as is the case in (26). Blends also tend to preserve the main stress of at least one of their source words (the stressed syllables in examples 24– 27 are underlined). In some cases (24, 26), the main stress of both source words is kept, or, when this is not possible, most often the stress of the second source word is preserved (25).
(24) bordinary < boring + ordinary (25) dramality < drama + (re)ality (26) predictionary < prediction + dictionary (27) prowebstinate < procrastinate + web (Beliaeva [2019])
23 If we now reconsider more exotic blend forms exemplified in Section 2.1, we can observe that peripheral cases of blending such as the ones involving intercalation or a combination of more than two elements are highly marked and thus attention- catching. According to Renner [2015], structural complexities such as intercalation in (16) and (19) or the presence of three or more structural elements in (19–20, 22–23) are factors that contribute to the playfulness of blends. Thus, more typical blends which can be regarded as more well-formed, may at the same time fulfill the word play function less successfully. However, note that blends that look not typical in regard to their linear structure, such as a central replacement blend in (27), or blends that do not include word beginnings in (17–18), nevertheless conform to the same phonotactic restrictions, that is, preserve the prosodic contour (the number of syllables and the main stress) of the longest of their source words. This may be related to the recognisability of the source words which, as shown in Gries [2006, 2012] and in Beliaeva [2014], is a key factor affecting blend formation. As gathered in Renner [2015], increased difficulty in recovering the constituents of the blend makes blends more playful. However, for the word play to occur, the constituents of blends have to remain potentially recognisable. Increased complexity of form involves lower recognisability of constituents, and therefore some compensatory formal features increasing recognisability are necessary. The preservation of the prosodic contour of one of the source words (particularly the one which loses its initial part) can be such a compensatory feature, being an essential prerequisite for recognisability, as shown in Gries [2012], Arndt-Lappe & Plag [2013] and Beliaeva [2016]. Therefore, increased playfulness may be associated with atypical form, but not all formal deviations are equally plausible for enhancing playfulness. We will return to this apparent paradox after considering the criteria of well-formedness of blends in relation to the possibility of recurrent use of their constituents.
3. Predictability of blends and its interplay with the productivity of splinters
3.1. Analysis of splinter productivity: Rationale
24 A distinguishing feature of blends is that, while being blended, words are split at points different from morph boundaries. Occasionally, counterexamples can be found. Thus, in (28–29) a compound is blended with a monomorphemic word so that the latter
Lexis, 14 | 2019 15
replaces either the first (28) or the second (29) compound constituent, and the structure of the resulting blend formally coincides with that of a typical compound.
(28) cookprint < cook + (foot)print (Word Spy) (29) passthought < pass(word) + thought (Word Spy)
25 More often, however, one or both source words of blends are cut to what is commonly referred to as ‘splinters’ (Adams [1973], Cannon [1986], Kemmer [2003], López Rúa [2004]). The boundaries of splinters are determined by factors other than morph boundaries. On the one hand, the switch point in blends normally falls on the syllable boundary or the boundary between onset and rime (see Kubozono [1990], Kelly [1998], Plag [2003] and Bauer [2012] for more detailed discussion), which is in agreement with phonotactic restrictions discussed in Section 2.1. On the other hand, the splinter length is subject to recognisability of the source words, that is, splinters tend to be long enough for the words they originate from to remain recognisable. In particular, as shown in a much-cited study by Gries [2006], the switch point in blends is usually close to the ‘recognition point’ of their source words, that is, a point at which the word is distinguished from other words of language that begin or end with the same letter / phoneme string. If we also take into consideration the tendency of blends to maximize overlap between their source words, as mentioned in Section 2.1, it turns out that splinters are the outcome of a complex interaction of phonotactically and cognitively relevant factors. Many splinters are tailored for one particular blend they appear in. However, a notable number of splinters are subsequently used to form further blends. With regard to such cases, Kemmer [2003] concluded that blending can give start to a productive process the result of which will be a lexical family and, eventually, a bound morph.
26 It is not surprising that splinters can be one-off formations only existing in the blends they appeared in, given the complex interplay of factors involved in their forming to suit a particular environment. The conditions under which splinters are used recurrently to give rise to a family of blends are of particular interest. Building on the investigation in Bauer, Beliaeva & Tarasova [2019], this section will focus on the analysis of recurrent use of splinters in contemporary English language. The term ‘splinter’ will be used throughout the section, although note that some researchers have referred to similar units as combining forms (Lehrer [1998], Warren [1990]). The analysis in Bauer, Beliaeva & Tarasova [2019] included estimating productivity of splinters using different quantitative measures such as type frequency as an estimate of profitability, and the ratio of the number of hapax legomena to the token frequency as an estimate of potential productivity (see Baayen [1992, 2009] for a detailed discussion of the methods and Bauer, Beliaeva & Tarasova [2019] for a discussion of applying the methodology to splinters). Different quantitative measures, however, induced different estimates of splinter productivity, and provide little ground for explaining the observed tendencies in the data: […] such models will be of little value without thorough understanding of what the measures actually represent. [Bauer, Beliaeva & Tarasova 2019: 65]
3.2. Data
27 The present study aims to evolve on findings in Bauer, Beliaeva & Tarasova [2019] regarding factors pertinent to the productivity of splinters, and to set grounds for
Lexis, 14 | 2019 16
developing relevant quantitative analyses. The data for this analysis come from a set of novel formations containing splinters that was collected in 2017–2018 for Bauer, Beliaeva & Tarasova [2019]. The first stage of data collection involved selecting splinters which a) have been used to form novel blends, and b) have been used to form more than one blend. These selection criteria indicated that the splinters had potential for productive use, with regard to criteria of potential productivity discussed, for example, in Baayen [1992] and Arndt-Lappe & Plag [2013]. Therefore, novel blends were selected from a collection of blends attested in COCA or in online dictionaries and neologism databases (Word Spy, Urban Dictionary and The Rice University Neologisms Database) no earlier than in 2000 (as confirmed by a Google search within a specific timeframe, see also Beliaeva [2014] for a detailed discussion). Out of over 500 novel blends in the original collection (Beliaeva [2014]), pairs or sets of blends with identical initial or final splinters were identified. As a result, 22 splinters were selected as having potential for recurrent use. Out of these splinters, 11 were initial, such as adver‑, originally the initial part of advertisement (30), and 11 were final, such as ‑cation, originally the final part of vacation (31).
(30) adver(tisement): adverblasting (Urban Dictionary), advergame (Word Spy) (31) (va)cation: staycation (Urban Dictionary), mancation (Urban Dictionary)
28 The full list of initial splinters, with indication of the words they originate from, is provided in (33), and the full list of final splinters is provided in (34), the parts of the original source words that have been truncated are in parentheses.
(33) adver(tisement), alterna(tive), digi(tal), edu(cation), fabu(lous), fem(ale), fin(ancial), loca(l), robo(t), virt(ual), yester(day) (34) (su)burb, (va)cation, (edu)cation, (in)flation, (or)gasm, (ep)isode, (para)noia, (an)orexia, (no)stalgia, (God)zilla, (Mo)zilla
29 Note that some of the final splinters in (34) are homonymous, that is, some of attested novel formations ending in ‑cation are blends using the final part of the word vacation, others are blends using the final part of education. Likewise, either the word Godzilla or the name of an Internet browser Mozilla could be a source word of a formation ending in ‑zilla. For each particular case in this data set, the actual source word has been determined by analysing the meaning of the context. The analysis in Bauer, Beliaeva & Tarasova [2019] did not include homonymous splinters and was therefore restricted to 9 final splinters and an equal number of initial ones. The present analysis will include all 11 initial and 11 final splinters from the original data set.
30 In order to investigate the use of splinters in contemporary language and estimate their productivity, words starting or ending with particular splinters were extracted from NOW (Davies [2013–]), a corpus that currently contains about 8.5 billion words from a variety of web-based media from 2010 to the present. After excluding proper names and derivatives starting or ending with the same letter strings, the data set comprised a total of 3007 word types. Examples of formations with different splinters, alongside the number of types of these formations attested in NOW as of the time of collection are provided in Table 1.
Lexis, 14 | 2019 17
Table 1. Neological formations containing splinters
Splinter Examples Number of types in NOW
adver‑ advercasting, advergame, advertainment 15
alterna‑ alterna-history, alternarock, alternanthem 68
digi‑ digimind, digisafe, digishoppers 605
edu‑ eduloan, edu-speak, educhat 267
fabu‑ fabuhealth, fabulash, faburrito 21
initial fem(i)- femgamer, femicare, femedic 179
fin‑ fincorp, finspy, finvest 36
loca‑ locabranch, locanomics, locavore 28
robo‑ roboracers, robobortion, robo-voting 1180
virt‑ virtnet, virtuheart, virtools 41
yester‑ yestermonth, yester-regime, yesterworld 32
‑burb boomburb, urbanburb, cluburb 19
‑cation1 brocation, daycation, farmcation 71
‑(u)cation2 fooducation, musication, NETucation 9
‑flation foodflation, joyflation, slowflation 38
‑gasm eargasm, mindgasm, scorgasm 105
final ‑(i)sode appisode, operasode, twittersode 14
‑(a)noia chemonoia, powernoia, juvenoia 18
‑(o)rexia bleachorexia, brideorexia, manorexia 22
‑(o)stalgia lustalgia, snowstalgia, technostalgia 23
‑zilla1 barzilla, sisterzilla, optimizilla 198
‑zilla2 bugzilla, clonezilla, appzilla 18
3.3. Quantitative estimates of productivity and related shortcomings
31 It is noteworthy that not all items in the data set looked like typical blends, as described in Section 2.2. In fact, in most cases (2541 out of 3007 types, which is 84.5% of
Lexis, 14 | 2019 18
the data), a splinter was attached to a free morph or sometimes a morphologically complex word without any truncation of the latter. In such formations, initial splinters resemble prefixes (35), and final splinters resemble suffixes (36). For subsequent analysis, formations like those in (35–36) have been classified as similar to affixations.
(35) advergame < adver‑ + game, roboracers < robo‑ + racers (36) foodflation < food + ‑flation, snowstalgia < snow + ‑stalgia
32 A much smaller number of items in the data set (7.2%, or 217 out of 3007 types) had properties of well-formed blends. In such formations, initial (37) or final (38) splinters were concatenated with shortened words, which often involved some overlap, or are merged with other words so that they overlap with the splinter – the overlapping segments in (37–38) are in bold type. Note that while determining the length of the overlapping segments, only the recurrently used splinters were considered as constituents of such blends, rather than the original source words, e.g. alterna‑, rather than its source word alternative, was analysed as the constituent of the blend alternanthem in (38). However, if we considered it the result of merging alternative and anthem, this item would still have been classified as blend conforming to properties of typical blends discussed in Section 2.2. In this respect, the distinction between splinters and their original source words does not play a significant role for the present analysis. However, for analyses where it might be relevant (e.g. for determining whether or not the prosodic contour of the source word is preserved), the original source words were taken into consideration.
(37) robostitute < robo‑ + prostitute, femedic < fem‑ + medic (38) alternanthem < alterna‑ + anthem, lustalgia < lust + ‑stalgia
33 In some cases (39), splinters were attached not to a free morph but to items that answer the description of neoclassical combining forms as defined in Plag [2003]. Occasionally, combinations of two recurrently used splinters were found (40). Such cases are formally similar to examples in (35–36), the only difference being that splinters are attached to bound morphs, rather than free morphs. To account for this difference, items such as the ones in (39–40) have been classified as similar to combining forms, which comprised 121 types, or 4% of the data.
(39) robocide < robo‑ + ‑cide, mega‑ + ‑gasm (40) robogasm < robo + ‑gasm, locavore < loca‑ + ‑vore
34 In the remaining 128 formations (4.3% of the data set), splinters were attached to clippings (41) or acronyms (42). The items formed in such way can be classified as clipping compounds or in some cases marginal blends, but since some items, e.g. VODzilla in (42), could be placed into either category, such cases were classified as ‘other’.
(41) femfest < fem‑ + fest(ival), tech(nology) + ‑gasm (42) digiTV < digi‑ + TV, VODzilla < VOD (video on demand) + ‑zilla
35 Formations like those in (35–36), where splinters are attached to free morphs or morphologically complex words in an affix-like manner, are the most frequently observed in our data. This is true about all the splinters in the data set, as can be seen in Figure 2, which displays the number of types where each of the splinters under
Lexis, 14 | 2019 19
investigation is attached to the other constituent of a novel formation in one of the following ways: similar to an affix, similar to a blend constituent, similar to a combining form (CF in Figure 2), or other.
Figure 2: Different structural types of blends in relation to additive and subtractive word formation categories
36 Such observation in itself should not be surprising given that affixation is a much more widespread way of forming new words in English than blending or using combining forms. This is not to say that the splinters in our data set have become affixes (which would be a premature conclusion), but this distribution is in accordance with the general observations regarding English word formation, e.g. those in Plag [2003]. Moreover, as observed in Bauer, Beliaeva & Tarasova [2019: 64], splinters with higher type frequency also tend to attach to free morphs in the majority of cases. Using quantitative estimates of productivity, however, did not provide an adequate explanation for the observed tendencies in the data because different measures (type frequency, type/token ratio and the ratio of the number of hapax legomena to the number of tokens) provided different results [Bauer, Beliaeva & Tarasova 2019: 64–65].
3.4. Taking more factors into consideration
37 To further explore the use of splinters and to gain more insights into their potential to participate in word formation, it was decided to take into consideration not only the type of concatenation and type frequency, but also possible factors that have been shown to influence the formation of blends, that is, the degree of overlap and prosodic contour. Therefore, the items in the data set have been coded with regard to the following parameters: the position of the splinter (initial or final), the type of concatenation (affix, blend, combining form, or other), type frequency, overlap (the number of graphemes which overlap where the splinter is attached to the word or other item), and the prosodic contour (that is, whether or not the syllable number and the main stress of the formation coincides with those of its first source word, the second source word, both of them, or none of them).
Lexis, 14 | 2019 20
38 It is important to note that the degree of overlap and the prosodic contour were estimated with regard to the original source words of the splinters. Thus, the overlap in (43) was estimated as 2 because two last letters of the source word floor overlap with the first two letters of orgasm, which is the source word of the splinter ‑gasm. Similarly, the formation in (44) was classified as preserving the prosodic contour of its first source word fabulous, and the one in (45) as preserving the prosodic contour of its second source word vacation, even though the actual coining of these items could have included simple addition of splinters fabu‑ and ‑cation, respectively. Such coding principles were followed in order to avoid ungrounded judgements on the actual origin of the items in the data set, and also not to disregard potentially valuable information. In the process of coding for overlap and prosody, the type of concatenation was therefore ignored. Formations that had the same number of syllables and main stress position as both their source words (46) were classified as preserving the prosodic contour of both source words.
(43) floorgasm < floor + (or)gasm (44) fabulips < fabu(lous) + lips (45) braincation < brain + (va)cation (46) digizen < digi(tal) + (citi)zen
39 In accordance with this procedure, each formation containing splinters with type frequency of up to 100 was manually coded for type of concatenation, overlap, and prosodic contour. For higher type frequencies, a random sample of 100 types has been used for coding all the variables except for the numeric value of type frequency, which remained unchanged. Therefore, the final set amounted to 1419 data points.
40 A conditional inference tree (decision tree) analysis was used to estimate whether type frequency of splinters is related to other parameters listed above. A decision tree method is used to estimate a regression relationship between the output variable (in this case, type frequency of a splinter) and a set of predictor variables (type of concatenation, overlap, and prosody), as described in Hothorn, Hornik & Zeileis [2006]. If the algorithm rejects the hypothesis that there is no relationship between the dependent variable and any of the predictor variables, the variable which has the strongest association with the dependent variable is selected and the data set is split into two groups (branches of the conditional inference tree), which are significantly different from one another (p<0.05). Such partitioning is repeated until further splits of data are no longer justified. The partitioning can be stopped earlier in order to avoid overfitting the model to the data and to mitigate the interpretation of the results. In the present analysis, the maximal number of levels on which the decision tree can be split was set to three, that is, each group of data could be split no more than three times before reaching the final node. The results of the analysis are displayed in Figure 3. Each node of the decision tree in Figure 3 is titled by the variable which determined the split at that node, the number of data points in each of the resulting nodes n is shown on top of the nodes, and the distribution of type frequency in each node is displayed as a boxplot where the black horizontal line marks median type frequency in the respective subset of data.
Lexis, 14 | 2019 21
Figure 3: A conditional inference tree estimating the association between the type frequency of splinters, type of concatenation, initial (ini) or final (fin) position of the splinter, and prosody
41 The results displayed in Figure 3 show that formations in which splinters are concatenated with free morphs like affixes (left branch of the decision tree) differ significantly from other types of concatenation (right branch of the tree) in terms of type frequency. In particular, items which bear formal similarity to affixations are more frequent than blends, combining forms and other types of concatenations. Other factors that appear to have association with type frequency are initial (ini) or final (fin) position of the splinter and, particularly for initial splinters, the preservation of the prosodic contour of the source words. Thus, the largest node in the right part of the tree (n=456) is characterized with the highest type frequency (median=200), affix-like type of concatenation, initial position of the splinter, and the prosodic contour which does not resemble that of any of the source constituents (w0). Such formations bear the most formal resemblance to affixations, as exemplified in (47).
(47) eduportal < edu‑ + portal, advergame < adver‑ + game
42 Interestingly, both for affix-like concatenations, and for the remaining data set, prosody appears to be associated with type frequency of initial splinters, rather than final ones. This seems controversial, given that preserving the prosodic contour of the word that loses its beginning (that is, in the case of the present data set, the source word of the final splinter) was considered an important characteristic of blends (see Gries [2012], Arndt-Lappe & Plag [2013], Beliaeva [2014]). Moreover, the partition of the data in the decision tree suggests that formations using different types of concatenation also differ in terms of prosodic characteristics. This can be illustrated by a mosaic plot in Figure 4.
Lexis, 14 | 2019 22
Figure 4: A mosaic plot showing the distribution of data by the type of concatenation and prosody
43 In a mosaic plot, larger areas correspond to greater number of observations having certain characteristics, and darker areas represent greater deviations of the corresponding numbers from chance. In Figure 4, the type of concatenation is plotted against the horizontal axis, and the vertical axis displays the prosodic characteristics of items in the data set (w0=prosodic contour differs from both source words, w1=the prosodic contour of the first source word preserved, w2=the prosodic contour of the second source word preserved, w1w2=the prosodic contours of both source words preserved). As can be gathered from Figure 4, if the formation conforms to the criteria of blends (type=blend), it is more likely that it will preserve the prosodic contour of its second source word, which is not the case with other types of concatenation. Similar inferences can be deducted from studying the distribution of the data according to concatenation type and the number of overlapping letters, which is displayed in Figure 5. In particular, the majority of items in the data set do not have any overlap between the constituents, but blends differ from the rest of the data in this respect as nearly half of items in this group are characterized with some degree of overlap.
Lexis, 14 | 2019 23
Figure 5: A mosaic plot showing the distribution of data by the type of concatenation and the number of overlapping letters
44 The observed results indicate that the type frequency of formations in the data set is subject to different factors from those associated with well-formedness of blends. The results of the quantitative analysis also show that splinters which have greater type frequency at the same time bear the least resemblance to blend constituents, as the majority of types of neological formations including these splinters do not involve any overlap and do not preserve the prosodic contour of their source words. This may indicate that greater productivity can be associated with losing characteristics of blend splinters and acquiring characteristics of a bound morph, which is also in accordance with observations in Kemmer [2003]. Respectively, novel formations containing splinters that have shown signs of increasing productivity, are less likely to conform to the criteria of well-formedness of blends discussed in Section 2.2.
4. Conclusion: Prototypical blends as points on a cline between creativity and productivity
45 As has been extensively discussed in the literature, and as reiterated in Section 2.2, the formation of blends is to a large extent predictable. In particular, typical blends consist of two elements, one of which is the beginning or the whole of one word, another is the ending or the whole of another word. In addition, the blend constituents tend to overlap where they are merged together, and the overall prosodic contour of the resulting formation resembles that of one of its source words, typically the second one.
46 There are many examples of blends and similar formations that deviate from these criteria of well-formedness, but, more importantly, different directions of possible deviation can be distinguished. On the one hand, it is possible to come across less typical formations which, at the same time, can be characterised by increased punning nature and playful character, as discussed in Section 2. On the other hand, if blend
Lexis, 14 | 2019 24
splinters are used recurrently, they may be attached to other language units in such a way that the resulting formations resemble affixations rather than typical blends. From this perspective, the characteristics of typical blends can be regarded as the result of an interplay between factors determining creativity and productivity of the word formation pattern, as visualized in Figure 6.
Figure 6: Deviations from prototypicality in blends in relation to creativity and productivity
47 In conclusion, the predictability of blend formation and the characteristics of typical blends cannot be ignored in a study of blending. However, the criteria of well- formedness may change for splinters that are used recurrently. What induces recurrent use remains an open question but since any splinter is, at least potentially, subject to recurrent use, it is important to take into consideration factors that may be associated with it. As shown in the present study, recurrent use of splinters may also involve less overlap and greater prosodic flexibility. At the same time, many items containing recurrently used splinters bear characteristics observed in typical blends. Thus, many of the affix-like concatenations discussed in Section 3.4 still preserve the prosodic contour of one of their source words (as was shown in Figure 4, see also discussion in Bauer, Beliaeva & Tarasova [2019]). It may be the case that for certain splinters the preservation of prosodic contour and therefore prosodic restrictions on the items the splinters may be concatenated with are more relevant than for the others, and factors involved in this are worth investigating in future studies.
BIBLIOGRAPHY
ADAMS Valerie, 1973, An introduction to modern English word formation, London: Longman.
Lexis, 14 | 2019 25
ADAMS Valerie, 2001, Complex words in English, Harlow: Longman.
ALGEO John, 1977, “Blends, a structural and systemic view”, American Speech 52, 47–64.
ARNDT-LAPPE Sabine & PLAG Ingo, 2013, “The role of prosodic structure in the formation of English blends”, English Language and Linguistics 17,3, 357–563.
BAAYEN Harald R, 1992, “Quantitative aspects of morphological productivity”, in BOOIJ Geert & VAN MARLE Jaap (Eds.), Yearbook of Morphology 1991, Dordrecht: Springer, 109–149.
BAAYEN Harald R, 2009, “Corpus linguistics in morphology: Morphological productivity”, in LÜDELING Anke & KYTÖ Merja (Eds.), Corpus linguistics: An international handbook, Berlin: De Gruyter Mouton, 899–919.
BAUER Laurie, 1998, “Is there a class of neoclassical compounds, and if so is it productive?” Linguistics, 36,3, 403–422.
BAUER Laurie, 2012, “Blends: Core and periphery”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), Cross-disciplinary perspectives on lexical blending, Berlin: De Gruyter Mouton, 11– 22.
BAUER Laurie, BELIAEVA Natalia & TARASOVA Elizaveta, 2019, “Recalibrating productivity: Factors involved”, Zeitschrift Für Wortbildung / Journal of Word Formation 3,1, 44–81.
BELIAEVA Natalia, 2014, “A study of English blends: From structure to meaning and back again”, Word Structure 7,1, 29–54.
BELIAEVA Natalia, 2015, “Blends at the interface between compounding and clipping: Evidence from readers’ evaluations”, Neologica 9, 205–219.
BELIAEVA Natalia, 2016, “Blends at the intersection of addition and subtraction: Evidence from processing”, SKASE Journal of Theoretical Linguistics 13,2, 23–45.
BELIAEVA Natalia, 2019. “Blending in morphology”, in Oxford Research Encyclopedia of Linguistics, Oxford: Oxford University Press, https://oxfordre.com/linguistics/view/10.1093/acrefore/ 9780199384655.001.0001/acrefore-9780199384655-e-511
BRDAR-SZABÓ Rita & BRDAR Mario, 2008, “On the marginality of lexical blending”, Jezikoslovlje 9,1–2, 171–194.
CANNON Garland, 1986, “Blends in English word formation”, Linguistics 24,4, 725–753.
DEVEREUX Robert, 1984, “Shortenings, blends and acronyms”, Word Ways 17, 210–215.
DRESSLER Wolfgang U., 2000, “Extragrammatical vs. Marginal morphology”, in DOLESCHAL Ursula & THORNTON Anna M. (Eds.), Extragrammatical and marginal morphology, München: Lincom Europa, 1– 10.
FANDRYCH Ingrid, 2008, “Pagad, chillax and jozi: A multi-level approach to acronyms, blends, and clippings”, Nawa Journal of Language and Communication 2,2, 71–88.
FROMKIN Victoria A., RODMAN Robert & HYAMS H. Nina, 2012, Introduction to language, Melbourne: Centage Learning Australia.
GRIES Stefan Th., 2004, “Isn’t that fantabulous? How similarity motivates intentional morphological blends in English”, in ACHARD Michael & KEMMER Suzanne (Eds.), Language, culture and mind Stanford, CA: Center for the Study of Language and Information (CSLI), 415–428.
Lexis, 14 | 2019 26
GRIES Stefan Th., 2006, “Cognitive determinants of subtractive word-formation processes: A corpus-based perspective”, Linguistics 17,4, 535–558.
GRIES Stefan Th., 2012, “Quantitative corpus data on blend formation: Psycho- and cognitive- linguistic perspectives”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), Cross- disciplinary perspectives on lexical blending, Berlin: De Gruyter Mouton, 145–167.
HOTHORN Torsten, HORNIK Kurt & ZEILEIS Achim, 2006. “Unbiased recursive partitioning: A conditional inference framework”, Journal of Computational and Graphical Statistics 15,3, 651–674.
KELLY Michael H., 1998, “To ‘brunch’ or to ‘brench’: Some aspects of blend structure”, Linguistics 36, 579–590.
KEMMER Suzanne, 2003, “Schemas and lexical blends”, in CUYCKENS Hubert, BERG Thomas, DIRVEN Rene & PANTHER Klaus-Uwe (Eds.), Motivation in language: From case grammar to cognitive linguistics. A Festschrift for Gunter Radden, Amsterdam & Philadelphia: Benjamins. 69–97.
KUBOZONO Haruo, 1990, Phonological constraints on blending in English as a case for phonology- morphology interface. in BOOIJ Geert & VAN MARLE Jaap (Eds.), Yearbook of Morphology 3, Dordrecht: Springer, 1–20.
LEHRER Adrienne, 1998, “Scapes, holics, and thons: The semantics of combining forms”, American Speech 73, 3–28.
LEHRER Adrienne, 2007, “Blendalicious”, in MUNAT Judith (Ed.), Lexical creativity, texts and contexts: The morphology/stylistic interface, Amsterdam & Philadelphia: Benjamins. 115–133.
LÓPEZ RÚA, Paula, 2002, “On the structure of acronyms and neighbouring categories: A prototype- based account”, English Language and Linguistics 6,1, 31–60.
LÓPEZ RÚA, Paula, 2004, “The categorial continuum of English blends”, English Studies 85,1, 63–76.
LÓPEZ RÚA, Paula, 2012, “Beyond all reasonable transgression: Lexical blending in alternative music”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), Cross-disciplinary perspectives on lexical blending, Berlin: De Gruyter Mouton, 23–34.
MARCHAND Hans, 1969, The categories and types of present-day English word-formation: A synchronic- diachronic approach (2nd ed.). München: Beck.
MATTIELLO Elisa, 2013, Extra-grammatical morphology in English: Abbreviations, blends, reduplicatives, and related phenomena, Berlin: De Gruyter Mouton.
MERWIN Hugh, 2013, May 9, “Introducing the cronut, a doughnut-croissant hybrid that may very well change your life”, Grub Street, http://www.grubstreet.com/2013/05/dominique-ansel- cronut.html
NASH Rose, 1970, “Spanglish: Language contact in Puerto Rico”, American Speech 45,3/4, 223–233.
PLAG Ingo, 2003, Word-formation in English, Cambridge: Cambridge University Press.
QUIRK Randolph, GREENBAUM Sidney, LEECH Geoffrey & SVARTVIK Jan, 1985, A comprehensive grammar of the English language, Harlow: Longman.
RENNER Vincent, 2006, Les composes coordinatifs en anglais contemporain, PhD dissertation, Université Lumière Lyon 2, Lyon.
RENNER Vincent, 2015, “Lexical blending as wordplay”, in ZIRKER Angelika & WINTER-FROEMEL Esme (Eds.), Wordplay and metalinguistic / metadiscursive reflection: Authors, contexts, techniques, and meta- reflection, 119–133.
Lexis, 14 | 2019 27
RONNEBERGER-SIBOLD Elke, 2010, “Word creation: Definition—Function—Typology”, in RAINER Franz, DRESSLER Wolfgang U., KASTOVSKY Dieter & LUSCHÜTZKY Hans Christian (Eds.), Variation and change in morphology Amsterdam: Benjamins, 201–216.
STEWART Heather, 2011, March 12, “Eurozone crisis: ‘Merkozy’ tries to ride to the rescue”, The Guardian: International Edition, https://www.theguardian.com/business/2011/dec/03/merkozy- euro-crisis-rescue
WARREN Beatrice, 1990, “The importance of combining forms”, in DRESSLER Wolfgang U., LUSCHÜTZKY Hans Christian, PFEIFFER Oscar E. & RENNISON John R. (Eds.), Contemporary morphology, Berlin: Walter de Gruyter, 111–132.
CORPORA
DAVIES Mark, 2008-, The Corpus of Contemporary American English (COCA): 400+ million words, 1990- present, https://corpus.byu.edu/coca/
DAVIES Mark, 2013-, Corpus of News on the Web (NOW): 3+ billion words from 20 countries, updated every day, https://www.english-corpora.org/now/
KEMMER Suzanne, 2003-, The Rice University Neologisms Database, http://neologisms.rice.edu/ index.php
Top Words for the first 15 Years of the 21st century and what they portend, 2015, http:// www.languagemonitor.com/global-english/top-words-for-the-first-15-years-of-the-21st- century-the-trends-they-portend/
Urban Dictionary, 2008-, http://www.urbandictionary.com
Word Spy, https://wordspy.com
ABSTRACTS
The formation of blends is predictable to a large extent. Moreover, it may give rise to productive constructions where part of a word that was once blended is used to form further blends such e.g. ‑cation (from vacation) in workcation, spa-cation, staycation, etc. However, not all blends demonstrate such potential and factors that can influence the productivity of blend splinters are worth investigating. This paper argues that blending, as a word formation process, can be driven by either factors enhancing predictability of the outcome or factors enhancing playful character of the formation, or both. Both qualitative and quantitative data analysis is used to investigate the relationship between well-formedness of the blends and their productivity potential.
La formation de mots-valises est prévisible dans une large mesure. De plus, elle permet de créer des constructions productives dans lesquelles une partie d’un mot qui a déjà été amalgamé est utilisée pour former d’autres amalgames tels que, par exemple ‑cation (de vacation) dans workcation, spa-cation, staycation, etc. Cependant, tous les mots-valises ne présentent pas un tel potentiel et il s’agit de prendre en considération les facteurs qui peuvent influer sur la productivité des fragments de mots-valises. Cet article avance l’hypothèse que la formation des mots-valises peut être dictée par des facteurs améliorant la prévisibilité du résultat ou par des facteurs renforçant le caractère ludique de la formation, ou les deux. L’analyse des données qualitatives et quantitatives est utilisée pour étudier la relation entre la forme des mots-valises et leur potentiel de productivité.
Lexis, 14 | 2019 28
INDEX
Mots-clés: mots-valises, amalgamation, linguistique de corpus, productivité, création des mots, formation des mots Keywords: blends, corpus linguistics, productivity, word creation, word formation
AUTHOR
NATALIA BELIAEVA Victoria University of Wellington, School of Linguistics and Applied Language Studies, New Zealand [email protected]
Lexis, 14 | 2019 29
Improving on observational blends research: regression modeling in the study of experimentally-elicited blends
Stefanie Wulff and Stefan Th. Gries
We thank Dylan Attal, Anna Bjorklund, Steven Critelli, Erica Drayer, Corinne Futch, Hali Lindsay, and Noah Rucker for their invaluable help with running the experiments, transcribing the recordings, and annotating the data.
0. Introduction
This study examines blending, that is “an intentional fusion of usually two words where a part of the first source word (sw1) – usually the beginning of sw1 – is combined with a part of the second source word (sw2) – usually the end of sw2 – where at least one source word is shortened and/or the fusion may involve overlap of sw1 and sw 2” [Gries 2012: 146]. Blending presents a curious case of word formation as it does not appear to be rule-governed as other derivational processes – usually, blending involves a conscious effort that involves word play, which often violates rigid morphological rules; it is less productive, yet at the same time, arguably more creative than most other derivational processes; and while it is superficially quite similar to other intentional word formation processes such as compounding, clipping, abbreviations, and acronyms, blending has received far less attention, maybe because the intricate interplay between orthography and pronunciation at play in blending is not a centerpiece of linguistic theory (see Gries [2012: 145]). Much of our previous knowledge of blends is based on observational data that, ultimately, may entail the risk of being opportunistic data samples. Therefore, in order to validate previous work based on these kinds of data, this paper compares results
Lexis, 14 | 2019 30
from an experimentally solicited sample of blends to the results from previously- collected observational data by studying the following three hypotheses: • Hypothesis 1: the shorter source word contributes more of itself to the blend;
• Hypothesis 2: sw2 determines the stress of the blend (more than sw1); and • Hypothesis 3: blending maximizes similarity between source words and blends. We first summarize previous research in particular with regard to the classificatory and descriptive questions they discussed and the kinds of data they used, before we turn to a description of our experiment and the data it provided (Section 2). Then, each of the three hypotheses is discussed (in Sections 3, 4, and 5 respectively) before we conclude (Section 6).
1. Previous research
Earlier research on blends focused mostly on classifying different types of blends, and how to distinguish blending from other word formation processes. One of the first such studies on blends is Pound [1914], who analyzed 314 blends. Pound defines a blend as “two or more words, often of cognate sense, telescoped as it were into one; as factitious conflations which retain, for a while at least, the suggestive power of their various elements” [Pound 1914: 1]. She argues that blends, which clearly fuse meanings and are consciously coined, should be considered distinct from analogical extensions and enlargements since these do not exhibit semantic fusion and are created unintentionally. Less clear to Pound is the distinction between blends and contractions [Pound 1914: 11]. Pound proposes qualitative labels such as literary coinages, speech errors, and conscious folk formations. In terms of structural analysis, however, Pound offers no groupings and even warns that “no very definite grouping seems advisable” since source words are combined in (apparently) unpredictable ways [1914: 22]. Algeo [1977: 48] defines blends as “a combination of two or more forms, at least one of which has been shortened in the process of combination.” This definition is based solely on blend structure and does not account for troublesome cases like meritocracy, which Algeo [1977: 54] defines as a derivative that combines with the form -ocracy. However, under Algeo’s own definition, meritocracy could be understood as a blend of merit and aristocracy. Algeo goes on to say that in some cases blends cannot be clearly distinguished from other derivational processes. For example, breadth could be understood as a blend (OE brede x length) but also as an analogical extension long → length : broad → x [1977: 51]. Differently from Pound [1914], Algeo offers two main classifications of blends. A first classification groups blends structurally into blends with phonemic overlap, blends with clipping, and blends with both phonemic overlap and clipping. A second classification distinguishes between syntagmatic blends, or telescope blends, of words that usually co-occur sequentially, like radarange (radar × range), and associative blends, which are blends whose source words are usually semantically linked in the coiner’s mind. Many researchers have articulated doubt that the formation of blends obeys any systematic rules – Bauer, for example, states that “in blending, the blender is apparently free to take as much or as little from either base as is felt to be necessary or desirable. […] Exactly what the restrictions are, however, beyond pronounceability and spellability is far from clear” [Bauer 1983: 225]. Nevertheless, several studies have made an effort to identify the cognitive determinants of blend formation. For example,
Lexis, 14 | 2019 31
Kelly’s [1998] analysis of 426 blends supports the idea that blending is predictable by revealing the following systematicities: • the first source words found in the blends are significantly shorter, significantly more frequent, and denote significantly more prototypical category members than the second source words; • the breakpoints of blends occur significantly more at syllable/word breaks than elsewhere (e.g., fool × philosopher → foolosopher, scum × company → scumpany, or sun × umbrella → sunbrella); furthermore, within-syllable breaks usually preserved the rime (e.g., breakfast × lunch → brunch, channel × tunnel → chunnel, or flight × plane→ flane). Gries [2004a] further examines the amount of information each source word contributes and the similarity of the source words to the blend. Building on Kaunisto [2000], Gries considers not only the graphemic, but also phonemic contributions of the source words to the blend, alongside length and medium as additional variables. The results of a loglinear analysis revealed that sw2 tends to be longer and contribute more of itself than sw1. Interestingly, his analysis reveals that the interaction between length and contribution is strong enough to not be affected by medium; that is, his results reveal a strong graphemic influence on blend formation, which is not observed in many other linguistic processes. Gries [2004b] investigates the degree of recognizability of the blend and the similarity of the sws to the blend. Examining the stress patterns of 614 blends with up to four syllables, Gries finds that sw2 plays a dominant role in determining the blend’s stress pattern. Gries [2012] distinguishes, if only heuristically, three stages of the blending process – (i) the selection of the source words to blend, (ii) the decision how to order them in the blend, and (iii) the decision how to split them up for the fusion – and shows that each of the stages exhibits distinct and significant patterns in their own right, but also when compared to (authentic and induced) error blends with regard to lengths and frequencies of source words, the similarities of source words to each other, and to the resulting blend (e.g., in terms of shared substrings, string-edit distances, and stress patterns). While these and other previous studies (see Renner et al. [2012]) have produced a wealth of results, they were all based on observational samples of blends collected by the researcher. This is potentially problematic in the same way that speech error collections often studied in the 1970s and 1980s are: it is not clear that the collection of the data is not affected by the ease with which certain blends can be perceived or memorized. In other words, not all blends occurring in real life – the population of blends, so to speak – have an equal chance of ending up in the researcher’s observational sample. A first experimental approach to follow up on the many observational studies was conducted by Arndt-Lappe & Plag [2013], who had 29 speakers of Irish English write and then pronounce blends in response to 60 written pairs of words; their source words systematically varied syllabic lengths and stress placements. They obtained altogether 1357 blend tokens from 107 ordered word pairs and largely corroborate existing observational studies (mostly focusing on Cannon [1986], Kubozono [1990], Gries [2004a-c], and Bat-El & Cohen [2006]): blends are typically as long as the longer source word, source words are often, but not always, split at constituent boundaries, and sw2 determines the stress of the blend more than sw1.
Lexis, 14 | 2019 32
In this paper, we will also discuss experimentally-obtained blends; in this first case study, the focus will be on validating previous observational research.
2. Data and methods
2.1. Experimental design
The source words to be used as stimuli came from four distinct semantic domains that represented plausible scenarios for intentional blending: fruit, vegetables, dog breeds, and car brands (see Appendix A for the full prompts participants were given). The specific source words selected for each domain were controlled for syllabic, graphemic and phonemic length as well as frequency to the best extent possible. Eight mono-, bi-, and tri-syllabic source words were rotated in each participant form. Monosyllabic source words had 3-4 graphemes/phonemes; bisyllabic source words had 5-7 graphemes and 4-6 phonemes (with the exception of kia with 3 graphemes and phonemes); and trisyllabic source words had 6-10 graphemes and 6-8 phonemes. The source words were presented in such a way that once all participants completed the experiment, mono-, bi-, and trisyllabic source words were blended together an equal number of times. Source word frequencies were obtained from the Corpus of Contemporary American English (COCA), a publicly accessible corpus covering spoken news reports and interviews, fiction writing, magazines, newspapers, and academic writing. Of the potential source words in each domain, the most frequent and the least frequent were selected. The resulting list of source words was the following: • fruit: banana, cantaloupe, cherry, grape, guava, plum • vegetables: bean, garbanzo, lentil, onion, potato, yam • dog breeds: chihuahua, lab, mastiff, poodle, pug, retriever • car brands: dodge, honda, jeep, kia, mercedes, pontiac To avoid potential priming effects, source words were never presented twice in a row as stimuli. Each participant saw an experimental form that contained 30 pairs of source words (15 pairs each from two out of the four semantic domains) and 30 filler items that served to shift participants’ attention from the blending task to a sufficiently dissimilar task. The filler items were simple math problems such as divisions and multiplications, rounding of numbers, and fractions. 12 unique experimental forms were created so that in a group of 12 participants, two participants saw source word pairs from the same two domains, yet in different order of presentation of sw1 and sw2.
2.2. Procedure
All experiments took place in the laboratory of Stefanie Wulff and were approved by the University’s Internal Review Board. All participants were college students enrolled at Stefanie Wulff’s university, and all were native English speakers between the ages of 18 and 25. A research assistant walked participants through the informed consent form and a participant information form that asked for personal information such as language background, age, and sex. Participants were then seated in front of a computer screen for the experiment. The experiment was conducted in two rounds. In Experiment 1 (E1), participants were presented with the stimuli and filler items on the computer screen and then asked to record their response in writing using pen and
Lexis, 14 | 2019 33
paper. In Experiment 2 (E2), a new group of participants were presented with the stimuli and filler items on a computer screen and then asked to articulate the stimulus or filler item out loud before recording their response in writing, and then to sound out their responses as well. To capture participants’ oral productions, the entire experimental session was tape-recorded. 72 students participated in E1, yielding 2,188 blends; 84 students participated in E2, yielding 2,520 blends (in both experiments, discarded responses included the participant saying “I don’t know” and repeating one or both source words without blending them). All written blends were copied into a spreadsheet, and all oral productions of source words and blends were transcribed using the CELEX phonetic alphabet [Baayen, Piepenbrock & Gulikers 1995].
2.3. Data annotation
Regarding the blend type, we determined for each grapheme/phoneme of the blend where its elements come from (we henceforth use the terms grapheme and letter interchangeably). For instance, consider Table 1 for our treatment of the well-known blend brunch. In this format modeled after Gries [2004c], the first two rows represent for each of the letters in sw1, breakfast, whether it is in the blend (lower row) or not
(upper row); the then next two rows do the same for sw2, lunch, just in the reverse order, which is so that the middle two rows highlighted in bold comprise the blend. The resulting annotation for BLENDTYPE is shown in the last row, namely for each letter in the blend which of the two source words – 1 or 2 – it is from. The current example highlights how our annotation identifies what is often considered the prototypical kind of blend – the beginning of sw1 followed by the end of sw2 – namely as a sequence of one or more 1s followed by a sequence of one or more 2s; in regular expressions, might one might summarize this as “1+2+”.
Table 1: Annotation of BLENDTYPE for breakfast × lunch → brunch
Letter slot 1 2 3 4 5 6 7 8 9
Letters from sw1 in the blend e a k f a s t
Letters from sw1 in the blend b r
Letters from sw2 in the blend unc h
Letters from sw2 not in the blend l
Annotation for letter BLENDTYPE 112222
This annotation can be extended to handle the maybe next most prototypical kind of blend, namely one that, around the point of fusion, involves overlap, i.e. graphemes or phonemes that occur in both source words, such as the l in fool × philosopher → foolosopher. These were marked with a 3, as shown in Table 2 for a blend from our data, potato × lentil → potatil.
Table 2: Annotation of letter BLENDTYPE for potato × lentil → potatil
Letter slot 1 2 3 4 5 6 7
Lexis, 14 | 2019 34
Letters from sw1 in the blend o
Letters from sw1 in the blend p o t a t
Letters from sw2 in the blend t i l
Letters from sw2 not in the blend len
Annotation for letter BLENDTYPE 1131322
Finally, there was a very small number of blends where the subjects coined a blend on the basis of the letters, but when they pronounced it, that blend contained a phoneme that was not represented in either source word, but instead resulted from the subjects ‘making phonemic sense’ of their graphemically-motivated creation; those were coded as 4; consider Table 3 and Table 4 for the letter and phoneme annotation of the blend jeep × honda → jenda, respectively. Additionally, for the oral responses, all source words and blends were also annotated for stress.
Table 3: Annotation of letter BLENDTYPE for jeep × honda → jenda
Letter slot 1 2 3 4 5
Letters from sw1 in the blend ep
Letters from sw1 in the blend j e
Letters from sw2 in the blend n d a
Letters from sw2 not in the blend ho
Annotation for letter BLENDTYPE 11222
Table 4: Annotation of phoneme BLENDTYPE for jeep × honda → jenda
Phoneme slot 1 2 3 4 5
Phonemes from sw1 in the blend ip
Phonemes from sw1 in the blend _
Phonemes from no sw in the blend e
Phonemes from sw2 in the blend n d %
Phonemes from sw2 not in the blend hQ
Annotation for phoneme BLENDTYPE 14222
Lexis, 14 | 2019 35
3. Hypothesis 1: the shorter source word contributes more of itself to the blend
In this section, we are revisiting the first hypothesis from above, which was first proposed by Kaunisto [2000] and then studied in, for instance, Gries [2004a-c].
3.1. Preparation of the data
In order to test Hypothesis 1, we needed the lengths of the source words in graphemes and phonemes as well as how much in percent they contributed to the blend. The graphemic lengths of the source words were straightforward to obtain from our master spreadsheet by just counting the number of characters for all source words. The contributions to the blends required a slightly more elaborate approach based on the blend lengths and their types as outlined above in Section 2.3. Based on that annotation, the contribution of
• sw1 to the blend was the number of 1s and 3s in BlendType divided by the length of sw1;
• sw2 to the blend was the number of 2s and 3s in BlendType divided by the length of sw2. 2 That is, for brunch (recall Table 1), the graphemic contributions of sw1 and sw2 are /9 4 and /5 respectively, for potatil (recall Table 2), the graphemic contributions of sw1 and 5 3 sw2 are /6 and /6 respectively, etc.
3.2. Statistical analysis
In the existing literature on this hypothesis, the lengths of the source words and their contributions were expressed in a ternary format. That means, comparisons were made between the source words of each blend to determine
• for lengths, whether sw1>sw2, sw1=sw2, or sw1 • for contributions to the blend as computed above, whether sw1>sw2 (i.e., whether sw 1 contributed more of itself than sw2), sw1=sw2, or sw1 Table 5: Cross-tabulation of source words’ lengths and contributions: observed frequencies (and Pearson residuals in parentheses) Contribution sw1 Lexis, 14 | 2019 36 sw1 sw1=sw2 120 54 (+9.98) 44 218 sw1>sw2 672 (+4.83) 62 260 994 Totals 1228 152 808 2188 However, the assumption of independence of data points that a chi-squared test relies on was already violated in the observational data. There, that violation was probably fairly inconsequential because the data comprised only a few blends that share certain source words; for instance, there were several blends with sex as sw1. But in the present experimental data, the amount of repeated-measurements structure of this type is of course much higher: all blends were created from the same set of source words, and every speaker contributed many data points. Thus, while the above results are suggestive, a better approach is needed. As an alternative, we adopted an ordinal mixed-effects modeling approach. For the dependent variable we first computed the following contribution percentage difference: contribution % sw1 – contribution % sw2. The resulting value ranged from -1 to +1: when it is high, sw1 contributes much more of itself to the blend than sw2; when it is low, sw1 contributes much less of itself to the blend than sw2; and when it is 0 or close to 0, both source words contribute about equally much. However, this set of values is very diverse (200 difference values with some less than 0.001 apart), many of them are only minimally different while at the same time meaning the same thing. Two differences of, say, 0.5 and 0.46 both mean sw1 contributes much more than sw2 – we do not need a linear regression to try to ‘explain’ that difference of 0.04 and would in fact not have much of a theoretical account at the level of quantitative resolution. Thus, we converted the difference values into a more useful ordinal response variable such that • if -1 < difference < -0.25, the response variable was set to “sw2 contributes more”; • if -0.25 ≤ difference ≤ 0.25, the response variable was set to “both contribute equally”; • if 0.25 < difference < 1, the response variable was set to “sw1 contributes more”. This response variable was then modeled as a function of each source word’s length (each as an orthogonal polynomial to the second degree to allow for curvature) and their interaction. As for the random effect structure, the only one that did not cause modeling problems consisted of varying intercepts for both sw1 and sw2 – additional varying intercepts for subjects exhibited very little variance in initial simple models and led to convergence problems with the fixed effects mentioned above. 3.3. Results The above model provides for a highly significant fit to the data (LRT=94.84, df=8, p<10-15), with the interaction of the two polynomials being significant as well (LRT=70.94, df=4, p<10-13), allowing for no obvious simplification to the model. However, the strength of the effect is small: Nagelkerke’s R2=0.05. While a higher R2 would have been desirable, the smallness of the value is not really surprising given that blend production is affected by many different and consciously manipulated factors, while we are testing only a single and very specific hypothesis here. Nevertheless, in order to be Lexis, 14 | 2019 37 safe, we computed two other mixed-effects models – one with the actual difference values as the response variable (e.g., the above 0.04) and one with a binary response variable (‘sw2 contributes more’ vs. ‘it does not’). While the numerical results differ, their implications with regard to Hypothesis 1 do not, which is why we proceed with our interpretation from what we considered to be the ‘best’ response variable. Given the nature of our model – an ordinal model with polynomials interacting – its interpretation on the basis of the numerical results is impossible. We therefore proceed on the basis of predicted probabilities of the three outcomes, but since we have two numeric predictors and three levels in our response variable, the resulting 3- dimensional graphs are instructive (and beautiful), but cannot be used in a non- interactive print medium. Instead, we represent the results in two 2-dimensional plots. Each plot in Figure 1 has the lengths of sw1 on the x-axis and the lengths of sw2 on the y-axis, and within the coordinate systems we are plotting 1s and 2s (when sw1 or sw2 is predicted/observed to contribute more of itself respectively) and “=” (when both are predicted/observed to contribute equally). In the upper panel, we plot the results predicted by the model, with greater font sizes indicating that the predictions are more confident (i.e., the predicted probabilities are higher). In the lower panel, we plot whether for each observed combination of source word lengths, the contribution of sw1 or sw2 was higher (plotting 1s and 2s respectively); empty slots in the lower panel mean that no such combination of source word lengths was observed in the data (e.g., we had no situation where both sw1 and sw2 were 7 characters long). Figure 1: Summary of the final model (graphemes, E1): predicted outcomes (upper panel) and observed outcomes (lower panel) Lexis, 14 | 2019 38 The overall complexity of the model notwithstanding, the results are interpretable and, in this case, fairly compatible with the simplistic chi-squared analysis, as is particularly clear from the lower panel: 1s (i.e. cases where sw1 contributes more) and especially big 1s are mostly found in the top left part of the plot, where sw1 is shorter than sw2, and the situation is the reverse for 2s. While the lower panel does not show any “=”s, it does show that many of the physically smaller 1s and 2s (i.e., when the distribution of the data is not clearly biased in favor of 1 or 2) are close to the main diagonal, where both source words are equally long. What about Hypothesis 1 for the phonemic contributions of source words in E2? The result of the initial exploratory chi-squared test for the phonemic lengths and contributions from E2 was extremely similar to that of E1: X2=194.7, df=4, p<10-10, V=0.1987, with the same three positive residuals only. For the same reasons as above, however, we proceeded with the ordinal mixed-effects model with the same fixed- effects predictors (just for the phoneme data in E2) and the same random-effects structure (this time, however, varying intercepts per subject could be included in the model without problems). This model, too, provides for a highly significant fit to the data (LRT=48.96, df=8, p<10-7), with the interaction of the two polynomials being significant as well (LRT=28.81, df=4, p<10-5), allowing for no obvious simplification to the model; however, the strength of the effect is even smaller than before: Nagelkerke’s R2=0.022. Again we computed two other mixed-effects models and again their results led to the same implications with regard to Hypothesis 1. Consequently, the visualization of the results in Figure 2 is the same as above. The model predictions in the upper panel are not particularly Lexis, 14 | 2019 39 instructive, which is not surprising given the very low R2, but the lower panel is a bit more informative: there are more and bigger 1s in the top left triangle (where sw1 is shorter than sw2) and there are more and bigger 2s in the bottom right triangle (where sw2 is shorter than sw1), which is indeed as expected. Figure 2: Summary of the final model (phonemes, E2): predicted outcomes (upper panel) and observed outcomes (lower panel) Lexis, 14 | 2019 40 In sum, the effects obtained from the experimental data are in the hypothesized direction – the shorter source word contributes more of itself to the blend – but they are noticeably weaker than they were in the observational data. In other words, while the previous results are supported, the present data also raise the specter that the convenience-sampling kind of approach that accounts for part of the observational data appears to amplify certain effects, maybe because the people who identified the blends unwittingly were more likely to notice formations as blends if they exhibited the hypothesized structure. 4. Hypothesis 2: sw2 determines BLENDSTRESS (more) 4.1. (Additional) Preparation of the data A first exploration of Hypothesis 2 consisted again of 3-dimensional cross-tabulation, namely cross-tabulating the four stress patterns of each sw1 and sw 2 (stressed: S, stressed-unstressed: SU, stressed-unstressed-unstressed: SUU, and unstressed-stressed- unstressed: USU) with the stress patterns of the blends provided by the participants in E2. However, the 12 stress patterns of the blends were quite Zipfian-distributed, which would be problematic for most kinds of categorical data analyses, which do not usually respond well to response variable with 12 levels, four of which are attested less than four times. At the same time, the four most frequent blend stress patterns not only accounted for nearly 91% of all tokens, but were also exactly the stress patterns exhibited by the source words. In order to describe the data best, we proceeded to do Lexis, 14 | 2019 41 both analyses. In what we will now call analysis1, we created a variable BLENDSTRESSWHENCE, which stated for each blend where it got its stress pattern from; for that we needed to distinguish four levels: • a level sw1, if the blend had sw1’s stress pattern, and sw2’s stress pattern was different; • a level sw2, if the blend had sw2’s stress pattern, and sw1’s stress pattern was different; • a level sw1sw2 if both sw1 and sw2 had the same stress pattern as the blend; • a level neither, if the blend had a stress pattern different from sw1 and sw2. This variable then became the response variable in a first statistical analysis discussed presently. However, in the other analysis, which we will now call analysis2, we ‘reduced’ the data by discarding the ≈9% of cases where the blend had a stress pattern that was neither that of sw1 nor that of sw2. 4.2. Statistical analysis As before, a 3-dimensional chi-squared test or a hierarchical configural frequency analysis of either data set would have been possible, but also probably problematic, given the repeated measures structure in this data set. Therefore, we opted again for a mixed-effects model, this time – given our response variables BLENDSTRESSWHENCE (in analysis1) and BLENDSTRESS (in analysis2) had four levels – Bayesian multinomial mixed- effects models. The predictors were SW1STRESS and SW2STRESS as well as their interaction, the random-effects structure consisted of varying intercepts of each sw1, each sw2, and each participant; our modeling parameters were four chains each with 2000 iterations (after a burn-in for each of 1000); these numbers may seem low, but see the convergence results below. 4.3. Results The models from both analyses converged just about perfectly (no R-hat values >1.01) and resulted in quite a good fit and accuracy. While, to the best of our knowledge, R2- values for this kind of model are not available, the classification accuracy of both models are quite good: 61.5% for analysis1, 66.7% for analysis2, which according to an exact binomial test is highly significantly better (p<10-96 and p<10-184 respectively) than the baselines of 40.75% or 36.8% (the frequency of the most frequent level of the response variables). Interpreting the results of such models is more difficult than those of ordinal models: A multinomial model with an interaction like ours would return four (levels of predictor 1) times four (levels of predictor 2) times four (levels of the dependent variable) = 64 predicted probabilities and/or 64 observed probabilities. We therefore proceeded as follows (described here first for analysis1): First, for all 16 combinations of the four levels of the two predictor variables SW1STRESS and SW2STRESS, we identified which level of BLENDSTRESSWHENCE had the highest predicted probability and what that predicted probability was. These are represented in the upper panel of Figure 3, which shows the levels of SW1STRESS in the ‘outer’ y-axis and the levels of SW2STRESS nested within those, with the predicted probability indicated by “×” on the x-axis with a small label on top of the “×” representing which level of BLENDSTRESSWHENCE is predicted for that combination of SW1STRESS and SW2STRESS. That means that the second row from the top indicates the Lexis, 14 | 2019 42 following: ‘When SW1STRESS is S and SW2STRESS is SU, then the blend is predicted to have the stress pattern of SW2 (with a probability of >0.7, 0718 to be precise).’ The crosses and labels are printed in blue when the blend did indeed exhibit the stress pattern of sw2 or of both sw1 and sw2, and else in red. For analysis2, we show the equivalent in the lower panel of Figure 3. Figure 3: Predicted probabilities of predicted outcomes of BLENDSTRESSWHENCE for all the data (upper panel) and predicted probabilities of predicted outcomes of BLENDSTRESS for the reduced data (lower panel) Lexis, 14 | 2019 43 The results are supportive of Hypothesis 2, but maybe not as strong as expected and maybe with a twist: Both panels show that, when SW1STRESS is S, sw2 – whatever its stress pattern – determines BLENDSTRESS, and when SW1STRESS is SU, then sw2 determines BLENDSTRESS unless sw 2 is monosyllabic. However, when SW1STRESS is SUU, sw 2 only determines BLENDSTRESS when it also is trisyllabic, and when SW1STRESS is USU, sw2 only ‘co-determines’ BLENDSTRESSWHENCE when it also does in analysis1 and only determines BLENDSTRESS in analysis2 when sw2 also is trisyllabic. While these results support that, on the whole, sw2 is a stronger determinant of BLENDSTRESS than sw1, part of the results is also compatible with the alternative (if only at times coincidental) account that the longer source word determines BLENDSTRESS. This is supported by the observation that trisyllabic sw1s determine BLENDSTRESS more than shorter sw1s. So, our above result of mostly sw2 determining BLENDSTRESS could be an artefact resulting from (i) BLENDSTRESS really being determined by the longer source word and (ii) the fact that previous studies have shown that sw2 is on average a bit longer than sw1 (e.g., Kelly [1998], Gries [2004c, 2012]). We therefore ran three separate models on the analysis2 version of the data: They all featured BLENDSTRESS as the response variable and the same random-effects structure as above, but the first one had the length difference in phonemes as a predictor, the second one the length difference in syllables between sw1 and sw2 as a predictor, and the third one the length difference in syllables as well as SW1STRESS and SW2STRESS and all their interactions as predictors. The results were unambiguous: All models converged but the classification accuracies of the first two did not even reach baseline Lexis, 14 | 2019 44 performance; the third model had a good classification accuracy of 67.1% (expectable since it featured the same two predictors that were already successful without the added length difference), but (i) that classification accuracy is not significantly better than the one above for the model without the length-difference predictor (pbinomial=0.3375) and (ii) a WAIC comparison showed that adding the length difference and its interaction to the model with ‘just’ SW1STRESS and SW2STRESS did not make the model reliably better (ELPD difference = -5.6, but with a standard error of 6.2). Therefore, this case study delivers results that are largely supportive of Hypothesis 2 and, thus, previous analyses based on the observational data. In addition, in our first multifactorial study of blend stress assignment, we also find that sw2’s dominance, so to speak, does not seem to be reducible to a length effect and does not benefit from being augmented with a length effect. 5. Hypothesis 3: blending maximizes similarity between source words and blends As discussed in much previous work, the similarity of blends to source words can be measured on a variety of dimensions as, for instance, in terms of stress pattern as in the previous section. This section focuses on similarity/distance in terms of graphemes (E1) and phonemes (E2). Here, we can adopt two perspectives: • similarity can be enhanced by picking two source words to blend that are already more similar to each other than random words are to each other; • similarity can be enhanced by blending the two source words in such a way that the resulting blend retains a high degree of similarity (and, thus, recognizability) to the source words. 5.1. The similarity of source words to each other The first perspective is in fact requires observational data because it can only be studied if one has a wide range of source word-blend combinations to look at. Previous work has confirmed that source words of blends are more similar to each other than random word pairs (or source words of complex clippings, for that matter; see Gries [2006, 2012]). However, we are also returning to this briefly here even with our experimental source words so as to offer at least an idea of how the source words we used compare to previous findings. To do that, we computed pairwise Levenshtein string-edit distances (SEDs) – i.e. the inverse of similarity – of the source words to each other for • the blends in our experimental data; • the blends in the latest version of Gries’s collection of observational data, which was last discussed in Gries [2012]; • 4708 pairs of randomly-chosen words, a number which corresponds to the number of experimental blends collected in both E1 and E2. In all these cases we computed both grapheme- and phoneme-based similarity. For 3 instance, the SED for channel and tunnel is /7 because one the longer of the two source words has seven characters and one needs three operations to get from channel to tunnel: deleting the c, replacing the h with a t, and replacing the a with a u. Then, we visually compared their empirical cumulative distributions, which are represented in Lexis, 14 | 2019 45 Figure 4. It is plain to see that the string-edit distances of words have relatively similar medians (of around 0.8 or 0.85 at y=0.5) and similar curves. Somewhat unsurprisingly, the source words of the blends from the observational data have the lowest distances – i.e. the highest degrees of similarity – but the reassuring finding is that our stimulus source words do not already behave very differently (in either direction). Figure 4: Ecdf plots for string-edit distances between words based on graphemes (upper panel) and phonemes (lower panel) 5.2. The similarities of source words to blends As for the second perspective and to measure how much the blending of the source words leads to a similar blend, we followed Gries’s [2012] general logic and computed for each blend an average Levenshtein string-edit distance (ASED) value, which means, we took the average of the SED between sw1 and the blend and the SED between sw2 and the blend. For channel × tunnel → chunnel, the ASED is 3/14=0.2143, namely the mean of 1 2 the SED of channel and chunnel ( /7) and the SED of tunnel and chunnel ( /7). We did this for • all experimental blends of our data, using graphemes for the blends in E1 and phonemes for those of E2; • all observational blends of Gries’s [2012] data, using both graphemes and phonemes; • all blends one could hypothetically generate from 6 pairs of source words (sw1s: strong, rich, television, flight, sloppy, Chevrolet; sw2s: powerful, handsome, armchair, suitcase, medical, Cadillac) while respecting phonotactic rules of English, meaning we did not include a hypothetical blend rich × handsome → rndsome. Then we fit a linear model to see how much the ASEDs – the similarity-preserving ways in which blends are formed from the source words – vary as a function of Medium (graphemes vs. phonemes), Type (experimental vs. observational vs. hypothetical/simulated) and the SEDs between the source words. The model revealed a significant three-way interaction between these predictors (p=0.016), which is represented in Figure 5. Lexis, 14 | 2019 46 Figure 5: The effect of SED on ASED for graphemes in the final model (upper panel) and the effect of Type:Medium on ASED in the final model (lower panel) The upper panel indicates that for graphemes, all three blend types behave the same: the more similar the source words are, the more similarly they also are jointly to the blend. This is reassuring because it confirms previous results based on observational blends, namely that blend creation involves this kind of using similarity to enhance word play and recognizability. At the same time, it is surprising that the mechanistically-created simulated blends, which by definition do not heed to this, reveal the same trend. An exploration of means does suggest, however, that as expected, the simulated blends scored lower on ASEDs than the other two kinds of blends. For the phonemes, the results are reassuring: the experimental blends behave just like the observational ones, and both are significantly different from the simulated blends. We did not include confidence intervals to reduce visual clutter, but the 95%-CI for simulated blends (phonemes) includes 0, reflecting that their similarity to the source words does not increase even as the source words become more similar to each other. All in all, we find that previous results based on the observational blends are supported. While there is one effect we cannot at present account for – the fact that simulated blends score as high on similarity between source words and blends as experimental and observational blends – this effect does not undermine the main point of this section, namely that the experimental blends pattern like the observational ones from prior studies. 6. Concluding remarks In sum, the results are first rather encouraging. While many studies, including several of Stefan Th. Gries, have proceeded using collections of blends that were often accrued under less-than-ideal sampling conditions, the results of our three case studies join most of those by Arndt-Lappe & Plag [2013] and lend credence to this kind of previous work. Section 3 showed that the shorter source word indeed contributes more of itself to the blend (using ordinal mixed-effects modeling); Section 4 showed that sw2 is indeed most influential in determining blends’ stress patterns (using multinomial Lexis, 14 | 2019 47 mixed-effects modeling); and Section 5 showed that blending attempts to increase similarity between source words and blends (using traditional linear modeling). That being said, we have also seen at least a bit of evidence that the observational data studied much in the past can, under certain circumstances, impart anticonservative results in the sense that effects appear stronger in the observational data than in the more controlled experimental data. The fear that this might happen motivated this study in the first place, but then also means that much more such ‘validational work’ needs to be done to determine which other results, if any, were amplified due to the nature of the observational data. One other conclusion to be drawn from this study certainly for us is a recognition of how difficult some of these analyses are even just from a methodological and statistical perspective. Even the controlled experimental data required not only an inordinate amount of transcription and error-checking, but also a data management/processing and statistical approach that go beyond much of mainstream types of analysis (of blends, but maybe also in much of linguistics in general). While it is possible to get some results from simple cross-tabulation and chi-square tests (as in Gries [2012] or Arndt-Lappe & Plag [2013]), once one wants to go beyond this and adopt the kind of analyses common in other contemporary corpus- and psycholinguistic studies, things become complicated very quickly. For instance, our case study of Hypothesis 3 first generated complete null results until we noticed that the source word similarities must be included as a control variable – only then did we see the more reasonable results reported here. Given the multitude of results that still await similar kinds of validation and the large number of factors that affect blend formation or at least need to be controlled for, blend researchers certainly have their work cut out for them. BIBLIOGRAPHY ARNDT-LAPPE Sabine & PLAG Ingo, 2013, “The role of prosodic structure in the formation of English blends”, English Language and Linguistics 17(3), 537-563. BAAYEN Harald R., PIEPENBROCK Richard & GULIKERS Leon, 1995, The CELEX lexical database (CD-ROM). University of Pennsylvania, Philadelphia, PA: Linguistic Data Consortium. BAT-EL Outi & COHEN Evan-Gary, 2006, “Stress in English blends: a constraint-based approach”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J. L. (Eds.), Cross-disciplinary perspectives on lexical blending, Berlin: Mouton de Gruyter, 193-212. CANNON Garland, 1986, “Blends in English word-formation”, Linguistics 24(4), 725-753. GRIES Stefan Th., 2004a, “Shouldn’t it be breakfunch? A quantitative analysis of the structure of blends”, Linguistics 42(3), 639-667. GRIES Stefan Th., 2004b, “Isn’t that fantabulous? How similarity motivates intentional morphological blends in English”, in ACHARD Michel & KEMMER Suzanne (Eds.), Language, culture, and mind, Stanford, CA: CSLI, 415-428. Lexis, 14 | 2019 48 GRIES, Stefan Th., 2004c, “Some characteristics of English morphological blends”, in ANDRONIS Mary A., DEBENPORT Erin, PYCHA Anne, & YOSHIMURA Keiko (Eds.), Papers from the 38th Regional Meeting of the Chicago Linguistics Society: Vol. II. The Panels, Chicago, IL: Chicago Linguistics Society, 201-216. GRIES, Stefan Th., 2006, “Cognitive determinants of subtractive word-formation processes: a corpus-based perspective”, Cognitive Linguistics 17(4), 535-558. GRIES Stefan Th., 2012, “Quantitative corpus data on blend formation: psycho- and cognitive- linguistic perspectives”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J. L. (Eds.), Cross- disciplinary perspectives on lexical blending, Berlin & New York: Mouton de Gruyter, 145-167. KAUNISTO Mark, 2000, “Relations and proportions in the formation of blend words”, Paper presented at the Fourth Conference of the International Quantitative Linguistics Association (Qualico), Prague. KELLY Michael H., 1998, “To brunch or to brench: some aspects of blend structure”, Linguistics 36(3), 579-590. KUBOZONO Haruo, 1990, “Phonological constraints on blending in English as a case for phonology- morphology interface”, Yearbook of Morphology 3, 1-20. RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), 2012, Cross-disciplinary perspectives on lexical blending, Berlin & New York: Mouton de Gruyter. APPENDIXES Appendix A. Prompts for the blending task (1) You are a marketing agent for a fruit snacks company that has just come out with a series of new fruit snacks that combines flavors of two different fruits. Your job is to entice people to buy the products by creating clever product names that combine the two fruits together. Keep the order of the fruits in the name the same as you are given. Example: peach × apple → papple (2) You are an agricultural scientist trying to patent new types of vegetables containing genetic material from two different types of vegetables. Unfortunately, competitors are also trying to patent the same combinations. You must come up with creative names for your new vegetables to ensure that the patented names are unique. Keep the order of the vegetables in the new names the same as you are given. Example: lettuce × radish → lettish (3) You are a dog breeder trying to get famous by coming up with the most popular new breed. You’ve decided to breed several different types of dogs together. For each of the following pairs, come up with a catchy name for the new type of breed by blending the names of the two types of dogs together. Keep the type of dogs in the order they are given. Example: beagle × husky → busky (4) For each pair of words you’re given assume you’re in a new merger meeting between two automobile companies. You’re a marketing agent whose job is to blend the Lexis, 14 | 2019 49 names of the car brands together in order to come up with a clever new car brand name. Keep the names in the order that you’re given. Example: Chevrolet × Cadillac → Chevradillac ABSTRACTS In this paper, we discuss the results of a blend production experiment and how it relates to previous research that was nearly exclusively based on observational data. Specifically, we study three different findings from published research, namely that (i) the shorter source word contributes more of itself to the blend than the longer source word, (ii) source word2 determines blend stress (more than source word1), and (iii) blending maximizes similarity between source words and blends. Using statistical techniques so far not employed in research on blends, we show that most findings from observational data regarding the three hypotheses studied are supported, but also occasionally tampered down. Cet article analyse les résultats d’une étude expérimentale de productions d’amalgames et la façon dont ils diffèrent ou non de ceux d’études antérieures fondées sur des données d’observation. Plus précisément, nous analysons trois conclusions tirées de recherches déjà publiées, à savoir : (i) le mot source le plus court contribue pour une part plus significative à l’amalgame que le mot source plus long, (ii) le mot source2 détermine l’accentuation de l’amalgame (plus que le mot source 1), et (iii) le processus d’amalgamation tire au maximum partie de la similarité entre les mots sources et les amalgames produits. Nous avons eu recours à des techniques statistiques non employées jusqu’à présent pour l’étude du processus d’amalgamation, afin de démontrer que la plupart des conclusions tirées des données d’observation quant aux trois hypothèses ci-dessus sont confirmées, mais doivent également parfois être modulées. INDEX Keywords: experimental blend production, observational blend collection, source word lengths, similarity, regression modeling Mots-clés: production expérimentale d’amalgames, collecte d’amalgames d’observation, longueurs des mots sources, similarité, modèle de régression AUTHORS STEFANIE WULFF University of Florida & UiT The Arctic University of Norway [email protected] STEFAN TH. GRIES UC Santa Barbara & JLU Giessen [email protected] Lexis, 14 | 2019 50 A corpus-based analysis of new English blends Mattiello Elisa Introduction [A]ll speech, smooth as well as blunderful, can be and must be accounted for essentially in terms of the three mechanisms… analogy, blending, and editing. [Hockett 1967: 935] 1 Lexical blending is a widely discussed topic in morphological literature. It has recently attracted the attention of several scholars, who have focused on 1) the role of prosodic structure in the formation of new English blends [Arndt-Lappe & Plag 2013], 2) prototypical blend features [Bauer 2012] and regularities in blend formation [Mattiello 2013], 3) the relationship between blend structure and meaning [Beliaeva 2014], and 4) the interpretation and acceptability of new lexical blends [Connolly 2013]. Lexical blending has also been studied from a cross-disciplinary perspective, in a volume collecting data from typologically different languages [Renner, Maniez & Arnaud 2012]. 2 This paper investigates new lexical blends in English, their frequency, pragmatic contexts, and functions, as well as the emergence of new splinters in their formation. In particular, the paper analyzes a collection of 245 blends from quantitative and qualitative viewpoints. The approach is both data-driven and corpus-based. For the analysis, the study uses lexicographic information drawn from the online version of the Oxford English Dictionary [OED2-3], as well as corpora of English, i.e. Corpus of Contemporary American English [COCA] and News on the Web corpus [NOW]. The data has been collected through an advanced search in the OED and covers a time span (1950-2010) which testifies to the current relevance and increasing incidence of the blending process as a word-formation mechanism for the creation of new words in English. The OED indeed shows that the number of new entries created by blending doubles in intervals of fifty years, namely, 33 instances in 1800-1850, 65 in 1850-1900, 147 in 1900-1950, and 246 in 1950-2000, as shown in Figure 1. Lexis, 14 | 2019 51 Figure 1: Advanced search results for ‘blend’ in the OED 3 The database of the study includes both “nonce formations” [Bauer 1983: 45; or “occasionalisms”, Chanpira 1966 in Dressler & Tumfart 2017: 155-156], which are coined for specific textual/stylistic purposes, and more stable neologisms, which are already or are likely to become a permanent part of the English lexicon (cf. “lexicalization” and “institutionalization” in Brinton & Traugott [2005: 45]). However, the attention is especially focused on neologisms, which from a quantitative viewpoint are more frequent and, from a qualitative viewpoint, can better illustrate the contexts motivating blending and the functions that new English blends can fulfil. 4 The goal of the study is fourfold: 1. First, it aims at identifying the contexts/registers which favour the formation of blend words, ranging from slang/colloquial registers (e.g. bromance ← bro + romance) to specialized domains. While the semantically hybrid nature of blending has been stressed in the literature – e.g., it is used to form names for unions, alloys, companies, etc. [Thornton 1993, 2004; Renner 2006; Bauer 2012] – and this process has often been regarded as a mechanism used to gain our attention in the media and advertising [Lehrer 2007; Ronneberger-Sibold 2010], its relevance to specialized vocabulary has not been adequately remarked hitherto. Pertinent domains for blends include, for instance, business (adhocracy ← ad hoc + bureaucracy), economics (stagflation ← stagnation + inflation), electronics (rectenna ← rectifying + antenna), technology (phablet ← phone + tablet), etc. 2. Second, this study addresses the question of whether blends are created with the intention of designating a new referent, thus filling in a conceptual/lexical gap, or to give a new name to an existing referent, as an act of economizing or creating a stylistic effect. For instance, the blend jeggings (← jeans + leggings) has been recently coined to designate ‘tight-fitting stretch leggings for women, styled to resemble a pair of denim jeans’, and it has even become a proprietary name in the United Kingdom. By contrast, vodkatini and surfari are merely shorter words for existing concepts expressed by compounds, i.e. vodka martini (‘a martini cocktail in which vodka is substituted for gin’) and surf safari (‘a journey made by surfers in search of good conditions for surfing’). The fact that only 4.9% of the blends in our database fall in the latter group confirms the necessity of blending as a word-formation mechanism used to label new complex concepts or objects, fusions, amalgams, and, more rarely, qualities or actions. 3. Third, the study focuses on a particular type of blending, called ‘attributive’ or ‘headed’, such as rockumentary (← rock + documentary), eatertainment (← eat + entertainment), and Lexis, 14 | 2019 52 Clintonomics (← Clinton + economics). Attributive blends exhibit an “endocentric relation” with their head [Bat-El 2006: 67], which therefore has a greater semantic weight than the first component (i.e. the modifier). 4. In particular, in this study, we will address the issue of attributive blends as possible schema model for new blends, with the second splinter (e.g. -umentary, -tainment, and -(o)nomics) as potential combining form, or secreted affix [Fradin 2000], for novel formations. 5 Issues 2 to 4 also question the boundaries between blending and clipped compounding [Beliaeva 2014], as well as between blending and frequent splinters, combining forms, or secreted affixes [Mattiello 2018]. 1. Theoretical background 1.1. Definition and classification of blending 6 Linguists generally agree on the definition of lexical blending as the merging of two (or more) lexemes into one involving partial loss of the phonological and/or graphic material of at least one of them, as in smaze ← sm(oke) + (h)aze, with partial drop of both source words. Frequently, there is an overlap between the source words, as in boatel ← boat + hotel, with overlap of the phonemes /əʊt/ which favours the fusion. Connolly [2013: 3] respectively calls these two types “substitution blends” and “overlap blends”. 7 A different terminological distinction is provided by Ronneberger-Sibold [2006: 155], who classifies blends according to their degree of transparency, from very transparent “telescope blends” (e.g. G. Amtsschimmelpilz ‘red tape fungus’ ← Amtsschimmel + Schimmelpilz) to completely opaque “fragment blends” (e.g. Cujasuma ‘a brand of tobacco’ ← Cuba + Java + Sumatra). Three-member blends are infrequent in English (an example is turducken ‘a poultry dish’ ← turkey + duck + chicken). 8 Finally, we can classify blends according to a semantic criterion, differentiating the ‘coordinate’ type (frenemy is both ‘friend’ and ‘enemy’) from the ‘headed’ or ‘attributive’ type (e.g. slimnastics is ‘gymnastics that slim you down’). In the headed type, the right constituent (gymnastics) acts as head and therefore carries a greater semantic weight than the left one (slimming). 1.2. Blends vis-à-vis other morphological categories 9 While different labels and classifications abound in the literature, scholars also disagree on the distinction between blends and other neighbouring morphological categories, especially clipped compounds (e.g. froyo ← frozen yogurt) and secreted affixes (e.g. -aholic ← alcoholic in shopaholic). It is often remarked that blending can be compared to compounding because it combines two (or more) base lexemes in order to form a new one [Bauer & Huddleston 2002: 1636; Gries 2004: 639]. However, whereas compounding is generally regarded as a regular, productive process, blending is viewed as “irregular and unpredictable” [Connolly 2013: 3]. This is mainly due to the fact that compounding combines words, whereas blending combines word parts [Kemmer 2003: 75] and its output is not regularly and transparently analyzable into morphemes [Bauer 1983: 234]. Lexis, 14 | 2019 53 10 This difficulty in identifying regularities in blends has led Laurie Bauer to list prototypical phonological, structural and semantic features as “defeasible constraints” for blends’ description, i.e. discriminating between the core and the periphery of blends, thereafter concluding that “the category is a fuzzy one” [Bauer 2012: 11, 21; cf. formal, semantic and syntactic properties in Cannon 2000]. Bat-El [2006: 66], instead, provides a narrow definition of blends, which according to her “refer only to cases where the inner edges are truncated” (e.g. Oxbridge ← Ox(ford) + (Cam)bridge, jazzercise ← jazz + (ex)ercise). This definition excludes from the category of blending forms where the right edges of two words are truncated (e.g. sitcom ← sit(uation) + com(edy)) or where only the first word undergoes truncation (e.g. mocamp ← mo(tor) + camp), which rather belong to the category of clipped compounds [Bat-El 2006]. In general, a clipped compound differs from a blend because it is attested as compound before being shortened. Thus, while *jazz exercise is not attested in this full form, only the blend jazzercise [1976] is, sitcom [1964] is shortened from the compound situation comedy [1953] and mocamp [1967] is a clipped compound from motor camp [1925].1 11 In this paper, we do not adopt Bat-El’s narrow definition, but rather extend blending to a broader category which also includes the ‘intercalative’ type (e.g. ambisextrous, where sex is intercalated within ambi(dex)drous) [Kemmer 2003: 72] and the type obtained from two word beginnings (e.g. modem ← mo(dulator) + dem(odulator)), although these are rarer than the prototypical type merging the beginning of one word with the end of another (e.g. gasohol ← gas(oline) + (alc)ohol). The only restriction for blends seems to be that “the beginning of a blend cannot be the end of a word, e.g. *glyson for ‘ugly person’.” [Lehrer 1996: 364] (however, cf. the clipped compound blog ← weblog). 12 Another fundamental distinction is between blending and combining forms (or secreted affixes). Fradin [2000: 46-47], for instance, has provided criteria to discriminate between these two morphological categories. Phonologically, combining forms are obtained by shortening the beginning or the end of a lexeme (e.g. -ware ← (soft)ware in freeware, or eco- ← eco(logy) in ecotourism), whereas there are numerous patterns in blend formation (as shown above). Semantically, while in blends the semantic content of the components is kept intact, in secreted affixation some semantic elements are kept and others discarded (cf. “secretion” vs. “abbreviation” in Warren [1990: 119]). Thus, boatel ‘boat which functions as a hotel’ is a blend, while shopaholic ‘compulsive shopper’ is a secreted formation obtained by discarding the semantic element ‘alcohol’ from alcoholic. Therefore, secreted affixing also involves abstraction – i.e. -(a)holic conveys the meaning ‘person addicted to the thing, activity, etc., expressed by the first element’ (e.g. shopping) – while blending does not. 13 The distinction between compounds, combining forms, and blends leads us to a more general tripartition which is elaborated within the framework of Natural Morphology (esp. in Dressler [2000]), i.e. “grammatical” vs. “marginal” vs. “extra-grammatical” morphology. In this framework, compounds like situation comedy belong to grammatical morphology because they are regularly formed according to word-formation rules and their outputs are fully predictable from their inputs (in this case, [situation]N + [comedy]N). Combining forms like -(a)holic instead belong to marginal (but still grammatical) morphology, in that they are non-prototypical (i.e. at the boundaries) of morphology [Dressler 2000: 6-7]. In particular, they are transitional between two subcomponents of morphology (i.e. derivation and compounding), depending on whether we consider combining forms to be bound or free morphemes. In the OED, for Lexis, 14 | 2019 54 instance, -(a)holic is labelled “suffix” and the outputs foodaholic, newsaholic, shopaholic, spendaholic, workaholic, etc. are regularly created from it. Finally, blending, like clipped compounding (the sitcom type), is a creative technique which Ronneberger-Sibold [2010: 201] includes among “intentional extragrammatical operations”, as opposed to “unintentional” ones occurring, e.g., during first language acquisition or in speech- errors. The use of extra-grammatical operations for the creation of new lexemes is termed “word creation” by Ronneberger-Sibold [2010: 201], as opposed to regular word-formation. Accordingly, blending is viewed as part of word-creation, rather than as a regular process of word-formation. For example, blends like smaze and boatel are dismissed from morphological grammar because their input does not allow a prediction of a regular output like rules do, or their output is only partially predictable on account of some sub-regularities or preferences for prototypicality [Mattiello 2013]. 14 However, a subtler distinction should be made between new blends that are single formations created after a model word – e.g., boatel is created after the model of motel (← motor + hotel) – and new blends that follow a schema model. Indeed, there are some blend “splinters” [Lehrer 1996, 2007] which recur frequently in novel coinages, thus showing a tendency towards regularity [Mattiello 2018] and productivity [Plag 1999; Bauer 2001; Bauer et al. 2013]. For instance, attributive blends such as eatertainment, irritainment, and shoppertainment exhibit a splinter -tainment (shortened from entertainment) which is also found in earlier docutainment, edutainment, and infotainment. 15 In the present study, new blends with recurring splinters such as -tainment are included within “paradigmatic morphology” [Bauer et al. 2013], in that they suggest an analysis in terms of paradigmatic substitution or analogy. In particular, while boatel belongs to “surface analogy”, based on a unique model, eatertainment and the like belong to “analogy via schema”, with a set of prototype words as model [Mattiello 2017; see “schema” in Köpcke 1993; cf. “schemas” and “subschemas” within Booij’s 2010 Construction Morphology]. In this study, the delimitation between productive and unproductive splinters in blending will be supported by a corpus investigation. 1.3. Contexts and recognition of new lexical blends 16 Another issue which will be addressed in this paper concerns the contexts where new English blends are created and employed, as well as their naming function, especially in specialized domains. In the literature, Thornton [1993: 148] has claimed that in Italian “[t]he formation of blends is scarcely productive”, mainly limited to designate entities such as companies, associations, unions, alloys, synthetic textiles, chemicals, and hybrids in general. She added that the creation of blends is favoured by the iconicity principle of Natural Morphology: i.e., the mixture of alloys, chemicals, etc. (signatum) is iconically reflected by the mixture in the process of formation of their names (signans). This favours the use of blends with a labelling function. 17 Similar conclusions are drawn by Ronneberger-Sibold [2006: 161] for German: “a strong formal amalgamation of the blended words can mirror a corresponding fusion of their referents”. For instance, substances consisting of several amalgamated ingredients are iconically named by amalgamated nouns, such as the blend smog ← smoke + fog. In her view, this iconicity is chiefly illustrated by brand names, which are generally created to impress listeners, and by pharmaceutical products, which often have scientific names reminiscent of the chemicals making them up. Ronneberger-Sibold [2010: 206-207] also Lexis, 14 | 2019 55 claims that, in general, word-creation is used in the realms of humorous literature – i.e. literature for children or texts for adults, whose aim is to amuse their listeners or readers – and of advertising, where the shape of a word can attract the attention of potential customers. 18 In a study on novel English blends, Lehrer [2007: 128] has similarly remarked that “[t]he commonest places for blends to occur are in product names, advertisements, newspaper and magazine headlines and titles”. According to her, the creation of blends is mainly motivated by the goal to call the reader’s attention to the product, news item, etc. and elicit his/her favourable response, such as remembering the product name, buying it, reading the news piece, etc. However, Lehrer [2007: 129] also notes the ephemeral nature of most blends, which are often intended to be “nonce forms – items produced for a specific context and occasion”. In connection with this, we assume that, while blends created in familiar, jocular or humorous contexts may be short-lived ad hoc creations, those formed in specialized contexts, such as pharmacy, biology, or information technology, are more stable formations that experts adopt for their efficiency (Language Economy Principle) and effectiveness (Iconicity Principle). 19 The demarcation line between ‘nonce’ blends (or occasionalisms) and proper ‘neologisms’, i.e. intended to enter the lexical stock of a language, is also related to the role of blending in lexical innovation and in the overall process of language change [Connolly 2013; cf. Milroy 1992; Brinton & Traugott 2005]. While the frequent emergence of new lexical blends in English (Figure 1) suggests that blending is a common process for the creation of new linguistic material (innovation), the proliferation among wider groups of speakers (adoption) might not be homogeneous for all innovative lexical items. Experiments have demonstrated that there is no general consensus of the “adoptability” of innovative blends by native speakers [Connolly 2013: 3]. 20 According to Connolly [2013: 12]: The accessibility of the meaning of an innovative blend appears to play some role in whether or not it is adopted, but the perceived prestige and utility of the form must also be taken into account. 21 From this claim, we gather that the spread and recognition of lexical blends are mainly motivated by three factors. First, new lexical blends are widely accepted if their source lexemes are easily accessible (recognizable) and their meaning straightforwardly assigned. Second, their adoption is favoured by prestige, that is new lexical blends that are indicative of a prestigious norm of speech will be more likely to be adopted than those that are not. However, a substandard unit may also become lexicalized because it is amusing, humorous, or sarcastic. Third, new lexical blends are especially adopted if speakers perceive the utility of the form, e.g., to cover a conceptual or lexical gap in the language, or to produce a stylistic or textual effect. 22 In this paper, quantitative data will be used to 1) discriminate between nonce blends and blend neologisms, and 2) identify the preferential contexts and domains where blends are created, recognized, and adopted. Section 2 explains how data was collected and analyzed. Lexis, 14 | 2019 56 2. Data collection and method 23 The data collected for this study was drawn from the online edition of the OED. For the selection of relevant data, the advanced search tool available on the OED platform was used. The parameters of selection were the formation process involved and the attestation date. First, the filter ‘blend’ in the etymology slot gave us 632 instances of blending in the overall dictionary (or, at least, those which have been labelled blends by lexicographers). Then, the entries were chronologically ordered from Early Old English to the present time. This ordering showed that a substantial distribution of English blends especially occurred after 1950 (see Figure 1), although there were no contemporary examples dated after 2010. The filter ‘1950-today’ restricted the set to 264 results, which appeared to be a representative collection of novel English blends for our goals. The collection was finally cleaned via close reading of each entry. Closer examination allowed us to exclude: • Abbreviations from phrases: e.g., Amex ← American Stock Exchange. • Forms with affixes or combining forms: e.g., poofteroo ← poofter + suffix -eroo, Neorican ← neo- + Rican. • Word parts: e.g., -bot ← robot, -rific ← terrific. Although some blends are obtained from these splinters (e.g. mobot ← mobile + robot, yogarific ← yoga + terrific), the blends which were not attested as separate entries in the OED were not included in the database, because they could not be retrieved systematically. • Words whose origin is only analogical: e.g., outro is analogically coined after reanalysis of intro as a complex word in + tro, not a blend of out and intro. • Words whose origin was uncertain: e.g., scuzz might be either an abbreviation from disgusting or a blend from scum and fuzz. 24 The final database consists of 245 English blends, including 209 nouns (85%), 32 adjectives (13%), and 4 (1.6%) verbs. Among the nouns, 48 are spelt with initial capital letters. In other words, 23% of the nouns are proprietary or proper names. These percentages correspond to a scale of different referents and functions in blending formation: Common nouns 65.7% > Names 19.6% > Adjectives 13% > Verbs 1.6% General denomination > Labelling > Description of qualities > Reference to actions/ events. 25 Hence, preliminary quantitative results suggest that the denomination/labelling functions of blends prevail over the description of qualities or reference to actions. Needless to say, these results are also influenced by the larger size of the syntactic category of nouns vs. other word classes (verbs, adjectives, adverbs) in the English lexicon. 26 In addition, lexicographic examination provided useful information about 1) the register of the selected blends (esp. ‘colloquial’/‘slang’ vs. specialized fields), 2) the status of the words (e.g. ‘nonce-words’, ‘temporary words’, only one ‘historical’), and 3) the connotation of the words, sometimes described as ‘derogatory’, ‘depreciative’, ‘humorous’, or ‘jocular’. 27 As for the methodology, a data-driven approach was combined with a corpus-based investigation. The corpora of English selected for the analysis are freely available at Lexis, 14 | 2019 57 Brigham Young University website (http://corpus.byu.edu), where they were created by Mark Davies: • Corpus of Contemporary American English [1990-2017, henceforth, COCA], which contains more than 560 million words with different genres of texts (spoken, fiction, popular magazines, newspapers, and academic) (last accessed April 2019); • News on the Web Corpus [2010-present, henceforth, NOW], which contains 6.9 billion words of data from web-based newspapers and magazines (last accessed April 2019). 28 Given the different size of the two corpora, token frequencies were normalized for the quantitative goals. Both raw and per million word frequencies were checked for all blends in the two corpora. However, in frequency assessment, some occurrences had to be excluded because of their irrelevance to a study on blends. For instance, the ambiguity of the words chugger, faction, shim, and wuss gave many occurrences as results, both in COCA and in NOW, but not all of them corresponded to the blends in our database from, respectively, charity + mugger, fact + fiction, she + him, and wimp + puss. Other items excluded from quantitative results were personal names or names of companies in the corpora (Picon, Uniterm, Swingle, Skitch), which obviously did not match with the nouns picon (← picture + icon), uniterm (← unit + term), swingle (← swinging + single), and the verb skitch (← ski / skate + hitch) in our database. Finally, the blend nerk (← nerd + berk / jerk) was not included in counts because, in COCA, it only occurs as an acronym from ‘Never Eat Road Kill’, and occurrences of melded were rejected because they were past forms of meld instead of adjectives/past participles. For all of these reasons, the above-mentioned blends were not considered to discriminate between nonce formations and well-established neologisms. 29 In Section 3, results from quantitative investigation (§ 3.1.) will be supported by a qualitative analysis of the data in its contexts of use (§ 3.2.). Analogical blends and recurrent splinters will be considered in sub-section 3.3. A general discussion of results will follow in sub-section 3.4. 3. Analysis, results, and discussion 3.1. Nonce words vs. neologisms 30 In this sub-section, a quantitative analysis of the new blends in our database is carried out in order to distinguish nonce blends from blend neologisms, and thus investigate how blending contributes to the process of lexicalization [Blank 2001: 1605-1606; Brinton & Traugott 2005: 41]. In general, both nonce formations and neologisms are considered innovations of a language. Specifically, a “nonce formation” is “a new complex word coined by a speaker/writer on the spur of the moment to cover some immediate need” [Bauer 1983: 45]. If it remains a single instance in the historical record, it is named “hapax legomenon” [Brinton & Traugott 2005: 45]. By contrast, if it “comes to be accepted by part or all of the speech community”, it becomes a “neologism” [Brinton & Traugott 2005: 45]. Besides becoming relatively independent of context, a neologism is lexicalized, i.e. accepted into the lexicon of a language, and institutionalized, i.e. spread in the language of a community and established as the norm. Lexis, 14 | 2019 58 3.1.1. Nonce blends 31 In our database, 25 examples of blending (10%) are only attested in the OED, but unattested in the two corpora of English explored. Table 1 reports a sample (ten instances) of such nonce blends (all nouns except one adjective), followed by their source words, date/meaning, and a contextualized example taken from the OED quotes. The blends are arranged in chronological progression. Table 1: Nonce blends in the OED Blend Source words Date/Meaning Example I don’t know any cozzer who [1950] a policeman; a cozzer (n.) copper + rozzer would have tackled us, mob- detective handed as we were. (OED, 1955) The ground’s too squoggy to tell squoggy (adj.) quaggy + soggy [1950] wet and miry whether they’re male or female. (OED, 1950) [1951] a lumbering The saccharine of false purity saccharhinoceros saccharine + person with an exuded from every pore of this (n.) rhinoceros excessively effusive saccarhinoceros advocate of manner virtue. (OED, 1951) [1964] a hallucinogen Their interest in psychedelics and which induces utopiates seems to have been utopiate (n.) utopia + opiate fantasies or visions of borrowed from the hippie a perfect existence subculture. (OED, 2004) There will be ‘advertiques’ such as advertisement + [1968] a piece of early advertique (n.) classic Coca-Cola signs and beer- antique advertising material logo items. (OED, 1994) computer + Their ‘computeracy’ will provide literacy [1969] knowledge of or the basis for subsequent training, computeracy (n.) (cf. the adj. skill in using retraining and the career changes computerate in computers which will become increasingly Table 2) necessary. (OED, 1981) The vestock can be described as a [1975] a clerical stock sort of black bib, with a hard vestock (n.) vest + stock that extends to the upstanding collar round at the waist front. (OED, 1975) [1985] a type of police The ‘hoolivan’ designed to detect van carrying trouble-makers in football crowds, photographic and hoolivan (n.) hooligan + van was unveiled at the Chelsea-Luton video equipment for match at Stamford Bridge last observing crowd night. (OED, 1985) behaviour Lexis, 14 | 2019 59 They ran a major awareness [1985] expenditure on campaign called Monergy in 1986, monergy (n.) money + energy energy which was Energy Efficiency Year. (OED, 1992) Biotechnologists, as the earliest [1987] a scientist who genetic + gengineers were called, had no gengineer (n.) works in the field of engineer idea that their work would prove genetic engineering so fruitful. (OED, 1987) 32 Although the blends in Table 1 are distributed between 1950 and 1987, it is not surprising that all of them are attested before 1990. This suggests that they have not survived more recent times. One of the blends (i.e. saccharhinoceros) is defined as a ‘nonce-word’ by the OED lexicographers themselves. 33 Table 2 instead reports a sample of ten blends for which either/both COCA or/and NOW display few occurrences, specified in the fourth column (raw/pmw frequency). 65 instances in our database (26.5%) belong to this category. Table 2: Nonce blends in COCA and NOW Source Raw/Pmw Blend Date/Meaning Example words Frequency [1959] favouring […] they have a legalitarian legislation as the legality / standing before the law and legalitarian best means of 1/0.00 legal + human conscience to pursue their (adj.) bringing about (NOW) egalitarian right to self-determination. greater social (NOW, 2017) equality [1962] lunar […] communitarian theorist exploration moondoggle moon + 1/0.00 Amitai Etzioni was calling lunar regarded as a (n.) boondoggle (COCA) obsession a “moondoggle”… frivolous waste of (COCA, 2008) money or time [1968] fervent or It also pioneered modern excessive acronymania: its divisions were acronymania acronym + enthusiasm for 1/0.00 AMX (the DSP), PMX (publisher (n.) mania the use of (NOW) exchange) and MMX (network acronyms or media exchange). (NOW, 2017) initialisms [1968] a full-sized, Still, he’s had some experience in functional replica 4/0.00 the replicar realm with his replicar (n.) replica + car of a vintage or (NOW) home-built Ferrari 250 GTO. classic car (NOW, 2013) Lexis, 14 | 2019 60 [1974] a Puerto Trio New York was formed in Rican native to or 1994 by Guillermo Colon (leader, inhabiting New second voice and second guitar), Newyorican New York + 1/0.00 York City creating the sound of traditional (n., adj.) Puerto Rican (NOW) [1978] of or Puerto Rican jibaro music with a relating to Newyorican urban feel. (NOW, Neoricans 2010) [1978] a (usually It’s also known for the Christmas free) promotional 1/0.00 Book, a magalogue filled with magalogue magazine + catalogue (COCA) extravagant his-and-her gifts, (n.) catalogue designed to 4/0.00 such as airplanes and submarines. resemble a high- (NOW) (COCA, 2007) quality magazine [1981] the practice of Herbicides can also be delivered chemigation chemical + applying 3/0.00 through irrigation systems (n.) irrigation chemicals to (NOW) (chemigation). (NOW, 2017) crops by means of irrigation water [1981] familiar Najib revealed that computerate computerate computer + with or skilled in 1/0.00 corporate ownership was merely (adj.) literate the use of (NOW) 10% and yet to achieve a target of computers at least 30%. (NOW, 2013) [1988] a proprietary name for a type of History of BSkyB: when the square + diamond-shaped 4/0.00 ‘squarial’ met the dish British Squarial (n.) aerial dish aerial for (NOW) Sky Broadcasting (BSkyB) was receiving satellite created… (NOW, 2011) television broadcasts [1995] a list of Volume 2 contains a lengthy webliography web + electronic works 3/0.01 bibliography and webliography (n.) bibliography relating to a (COCA) of additional resources. (COCA, particular topic 2012) 34 The blends reported in Table 2 are first attested in the 1950s-1990s in the OED, but they also occur more recently in COCA and NOW. However, these blends have been considered nonce blends because their normalized frequency is no higher than 0.00 (max. 1 occ. in COCA or 1-4 occ. in NOW). One of them (i.e. Squarial) has been labelled ‘temporary’ in the OED. 35 Other blends in our database display the same low frequency in corpora: e.g., the adjective sexploitative [1973] (← sex + exploitative) occurs only once in NOW, but the related noun sexploitation ‘sexual exploitation’ already existed in 1924 and has a current Lexis, 14 | 2019 61 raw frequency of 185 occ. (0.03 pmw) in NOW (see also the noun sexploit [1960] ‘sexual exploit’, 3/0.00 occ. in NOW). 36 Similarly, the adjective televangelical [1976] (← television + evangelical) occurs only once in COCA, but the occurrences of the nouns televangelism [1958] (30/0.05 in COCA, 78/0.01 in NOW) and televangelist [1973] (167/0.30 in COCA, 1,494/0.27 in NOW) are undeniably higher. The existence of word families of blends including derived nouns and adjectives is also symptomatic of the productivity of the process and of the stability of the pattern. 3.1.2. Blend neologisms 37 Blend neologisms are more difficult to ascertain. In order to be considered a lexicalized word, the new blend should display high raw/normalized frequency in corpora. Table 3 only reports the blends whose normalized frequency is higher than 0.10 pmw in at least one (but preferentially both) of the corpora checked (overall 42 instances/17% in our database).2 These are considered to be the ideal candidates for lexicalization/ institutionalization in English. Table 3: Blend neologisms in COCA and NOW Source Raw/Pmw Blend Date/Meaning Example words Frequency [1955] any of a number of humorously 214/0.38 There’s a wonderful marionette + grotesque glove (COCA) interactive “create your own Muppet (n.) puppet puppets and 2,833/0.52 Muppet” display… (COCA, marionettes (NOW) 2017) developed by Henson [1958] a synthetic steroid, Global effects of a synthetic C22H29FO5, which GR ligand Dexamethasone deca- ‘ten’ + 152/0.27 resembles (Dex) have been studied in dexamethasone hexadecadrol (COCA) cortisone in its cells treated with Dex prior to (n.) + methyl + effects and is used 852/0.15 or together with an cortisone as an anti- (NOW) inflammatory stimulus. inflammatory (COCA, 2017) agent [1960] a person whose physical tolerances or 347/0.61 It was a prototype, cyborg, cybernetic + capabilities are (COCA) cyborg (n.) self-repairing, able to train its organism extended beyond 5,972/1.09 own caretakers. (COCA, 2017) normal human (NOW) limitations by a machine Lexis, 14 | 2019 62 The CED’s influence and [1965] a state of membership declined in the the economy in 115/0.20 “stagflation” era of 1973 to stagnation + which stagnant (COCA) stagflation (n.) 1980, when the Business inflation demand is 1,095/0.20 Roundtable supplanted it as accompanied by (NOW) the leading spokesman for big severe inflation business. (COCA, 2017) The Republican base was 164/0.29 (Ronald) [1970] the always wary of George H.W. Reaganomics (COCA) Reagan + economic policies Bush. In 1980, he called (n.) economics of Reagan 322/0.06 Reaganomics voodoo (NOW) economics. (COCA, 2015) [1973] an evangelical We were saying Mark Burns is 167/0.30 preacher who the pastor, the televangelist televangelist television + (COCA) appears regularly with the Now Network, (n.) evangelist on television to 1,494/0.27 because we jumped the gun. preach and appeal (NOW) (COCA, 2016) for funds [1975] a very intelligent person; 60/0.11 She is a brainiac, wicked brainiac (n., brain + an expert (COCA) smart, or as the Scots would adj.) maniac [1976] very 736/0.13 say, wicked smart. (COCA, intelligent or (NOW) 2014) clever [1976] the Indian She knew Sweetie had pulled 292/0.52 film industry, every string he could to get Bombay + (COCA) Bollywood (n.) based in Mumbai her the audition with Hollywood (formerly 122,908/22.35 Bollywood’s top dance Bombay) (NOW) troupe. (COCA, 2017) Yes, the least – the least [2001] intimate 80/0.14 convincing bromance of all bro + and affectionate (COCA) bromance (n.) time is Donald Trump and romance friendship 3,355/0.61 Mitch McConnell. (COCA, between men (NOW) 2017) [2002] a type of 21/0.04 She was also the first woman swimsuit for burka + (COCA) to wear a burkini during the burkini (n.) women which bikini swimsuit portion of the event. covers the head 2,259/0.41 (COCA, 2017) and body (NOW) 38 The blends reported in Table 3 are distributed from 1955 to 2002. Their high frequency in corpora suggests a status as ‘neologisms’. The specialization of some of them – e.g., dexamethasone is used in Pharmacology, cyborg belongs to Cybernetics and thence to science fiction, stagflation to the field of Economics, Reaganomics to Politics – guarantees Lexis, 14 | 2019 63 that they have been accepted at least within their respective specialized communities of experts. The acceptability and spread of the other blends may be instead linked to their dissemination through the media: e.g., Muppet and televangelist are heard in television, Bollywood is linked to the cinema industry, and burkini has recently heated religious discussions in the news. 39 A more systematic examination of the registers, contexts, and functions of new blends in English can help us corroborate the impact of blending on the innovation and lexicalization processes. 3.2. Registers, contexts, and functions of new blends 3.2.1. Registers 40 In our database, the prevailing registers of blend lexemes include slang (i.e. language of a highly colloquial type, considered as below the level of standard educated speech) (7.3%) and colloquial (i.e. informal) language (7%), on the one hand, and specialized language and jargon (24.5%) on the other. • Slang: slang blends are often employed by speakers who intend to catch the reader/hearer’s attention with their colourful language, to be humorous, jocular, or otherwise by those who aim at mocking, offending others by using a sarcastic or derisive tone. Novel instances include: chugger / chugging (← charity + mugger / mugging), cozzer (← copper + rozzer), feminazi (← feminist + Nazi), gaydar (← gay + acronym radar), kideo (← kid + video), ragazine (← rag + magazine), skeeze (← skank + sleaze), skitch (← ski / skate + hitch), snarfle (← snarf + snaffle / snuffle), and ultraviolation (← ultraviolet + violation). Some slang blends display a derogatory character (e.g. himbo ‘an attractive but unintelligent young man’ ← him + bimbo, brainiac ‘a depreciative word for a very intelligent person’ ← brain + maniac), or are even taboo words (fugly ‘very ugly’ ← fucking + ugly, Masshole ‘a term of contempt for a native of the state of Massachusetts’ ← Massachusetts + asshole). Others are mere innovative original forms used by the young (e.g. fantabulous ← fantastic + fabulous), or originated in the context of hip-hop music (e.g. crunk ‘exciting or fun’ ← crazy + drunk). • Colloquial language: colloquial blends have a familiar flavour, which is typical of informal registers. They include: blaxploitation (← black + exploitation), bromance (← bro + romance), buppie (← black + acronym yuppie), Cassingle (← cassette + single), gayby boom (← gay + baby boom), Mockney (← mock + Cockney), sexploit / sexploitative (← sex + exploit / exploitative), Trustafarian (← trust fund + Rastafarian), and wuss (← wimp + puss). Familiar blends, such as the form of address glam-ma (← glamour(ous) + grandma), add to other humorous nicknames, such as Socceroos (← soccer + kangaroo) for ‘the Australian national soccer team’, or Taffia ‘any supposed network of prominent or influential Welsh people’, from Taffy ‘familiar nickname for a Welshman’ and mafia. • Specialized language: specialized or domain-specific blends belong to a variety of different fields and sub-fields: ◦ Some belong to the areas of Economics (slumpflation ← slump + inflation), Business (adhocracy / adhocratic ← ad hoc + bureaucracy / bureaucratic, flexecutive ← flexible + executive, freemium ← free + premium, glocal ← global + local), and Law (legalitarian ← legality / legal + egalitarian). ◦ Several others are used in Politics (Clintonomics ← Clinton + economics, commentariat ← commentary + proletariat, dissensus ← dissent + consensus, hacktivism / hacktivist ← hack + activism / activist, militician ← military + politician, politicide ← political + homicide, Lexis, 14 | 2019 64 preferendum ← preference + referendum, selectorate ← selector + electorate, slacktivism / slacktivist ← slack + activism / activist, veepstakes ← veep + sweepstakes, priviligentsia ← privilege + intelligentsia). ◦ The scientific domain also provides a range of blends which pertain to Mathematics (diffeomorphism / diffeomorphic ← differentiable + homoeomorphism / homoeomorphic, flexagon ← flex + hexagon), Biology (algeny ← alchemy + gene, cybrid ← cytoplasmic + hybrid), Chemistry (dielectrophoresis ← dielectric + electrophoresis), Biochemistry (ubiquinone ← ubiquitous + quinone), Pharmacology (artemisinin ← artemisia + quinine, cephaloridine ← cephalosporin + pyridine, nitrofurantoin ← nitrofuran + hydantoin), Surgery (dermabrasion ← Greek δέρμα ‘skin’ + abrasion), and Genetics (biolistic ← biological + ballistic). ◦ The technological arena ranges from Computing (Centronics ← centre / central + electronics, Internaut ← Internet + astronaut, knowbot ← know + robot, netiquette ← net + etiquette, netizen ← net + citizen, Quotron ← quotation + electron, Usenet ← use + network) to Electronics (molectronics ← molecular + electronics, rectenna ← rectifying + antenna), from Cybernetics (cyborg) to Aeronautics (taileron ← tail + aileron). ◦ Domain-specific blends belong to Film/Television/Broadcasting (animatic ← animated + schematic, Britcom ← British + comedy, Japanimation ← Japan + animation, synthespian ← synthetic + Thespian), Science Fiction (gengineering / gengineer ← genetic + engineering / engineer, plasteel ← plastic + steel), Telecommunications (teletex ← telex + text), Architecture (Populuxe ← popularity + Fr. luxe), and Nautical (Panamax ‘a class of cargo ship’ ← Panama (Canal) + max). ◦ Minor domains include: Agriculture (chemigation ← chemical + irrigation, fertigation ← fertilizer + irrigation), Palaeography (expunctuation ← expunction + punctuation), and Geology (volcaniclastic ← volcanic + clastic). 41 Specialized contexts especially favour the creation of blends whose form mirrors their meaning. The iconicity of blends is especially evident in pharmacology, where names of substances (artemisinin, nitrofurantoin) refer to mixtures, combinations, fusions, but also in television and broadcasting, where an informercial (← information + commercial) is ‘an advertisement which promotes a product, service, etc., in an informative style’, or in publishing, where a magalogue is ‘a promotional catalogue designed to resemble a high- quality magazine’. 3.2.2. Contexts and functions/effects 42 Blends can be found in heterogeneous contexts and used in different circumstances, depending on their formal or informal nature, on the effects that they are meant to produce, and on the goals that speakers intend to achieve through them. 43 In magazine and news articles, for example, creative blends are used to catch readers’ attention and encourage them in reading the whole news items. Examples (1) and (2) are both extracts from magazines: (1) […] she sent a sext ‘sex + text’ from the presidential Twitter account. (COCA, 2016) (2) Ariela Barer will play Gert, “a purple-haired, bespectacled, contemporary riot grrrl” ‘grrr + girl’ with a strong sense of social justice. (COCA, 2017) 44 where sext refers to ‘a sexually suggestive message sent electronically’ and grrrl designates ‘a young woman perceived as strong and aggressive’. Readers will be particularly attracted by the originality of these expressions, and their attention will be Lexis, 14 | 2019 65 focused on the fact that the text is on sexually-related topics (sext), or that the character of the grrrl – with the interjection grrr jocularly alluding, by alliteration, to her name Gert – expresses feminine independence, fierceness, and liveliness. 45 Informal contexts favour blend creation with the aim of creating intimacy within a group, such as surfers: (3) […] they were interested and in 1965 organized a “surfari” ‘surf + safari’ to the west of Ireland. (NOW, 2010) 46 who use surfari to indicate ‘a journey made in search of good conditions for surfing’. 47 By contrast, specialized contexts favour blend formations for other reasons: e.g., in (4) Populuxe ‘popularity + luxe’ – American luxury for the masses – sent the wealthy looking to Europe… (COCA, 1996) 48 the blend Populuxe, derived from popularity and luxe ‘luxury’, perhaps with an allusion to pop art, has an aesthetic function. It indeed refers to ‘a style of architecture and home furnishings design characterized by futuristic shapes and ornamentation’. Hence, the embellishment and originality of the blend form, purposely coined to designate a revolutionary architectural style, may reflect the design of the buildings in this style. 49 In the fields of politics and economics, blends may refer to specific phenomena, events, or trends: (5) People thought Reaganomics ‘Reagan + economics’ were good, but Clintonomics ‘Clinton + economics’ is – is proving much better. (COCA, 1999) (6) Trumpflation ‘Trump + inflation’ becomes Slumpflation ‘slump + inflation’ as political uncertainty sees oil come to its senses and follow gold higher. (NOW, 2017) 50 In (5), for instance, the political blends Reaganomics and Clintonomics denote two U.S. Presidents’ (Ronald Reagan and Bill Clinton) economic policies, while in (6) Trumpflation is a novel term created after slumpflation [1974] referring to ‘a state of economic depression’. In specialized domains, the creation of new blends appears to be more regular: e.g., the target word Clintonomics [1992] is analogically coined after the model Reaganomics [1970], and the nonce form Trumpflation is modelled on earlier stagflation [1965] and slumpflation [1974], the latter anaphorically following the target in (6) (Mattiello 2017; more on analogy in § 3.3.). 51 Technological settings, such as computing and the Internet, can also accommodate novel blends, as in: (7) The unit introduces students to the structure of blogs, the concept of a world-wide audience, and general netiquette ‘net + etiquette’ policies. (COCA, 2012) (8) Kahn […] is the co-inventor of Knowbot ‘know + robot’ programmes – mobile software agents in the network environment. (NOW, 2015) 52 where netiquette alludes to ‘the behaviour of Internet users’ (cf. netizen) and knowbot denotes ‘a type of automated program or (Internet) software’ (cf. infobot). 53 Overall, in specialized environments, new blends have the function of creating cohesion among experts, as well as designating specific referents by means of efficient terminology. Lexis, 14 | 2019 66 54 The denomination/labelling function, however, is the most noticeable in our database, where new blends give names to language varieties, such as African-American English (Ebonics), or jocular accents imitating London Cockney (mockney): (9) He had no idea how to speak Spanish, didn’t speak Ebonics ‘ebony + phonics’, either. (COCA, 2017) (10) These scenes, and the mockney ‘mock + Cockney’ chants of “Who are ya?”, were widely derided as the work of wannabes posturing for the day. (NOW, 2018) 55 Blends are also used to refer to new types of music, such as soca, a variety of calypso (cf. rapso ← rap + calypso), or hip-hopera, combining the two genres: (11) Add in calypso and soca ‘soul + calypso’ music, and you have a recipe for success week after week... and year after year; the event dates to 2000. (COCA, 2011) (12) Years ago, someone (was it the Fat Boys?) attempted “hip-hopera” ‘hip- hop + opera’. (COCA, 1999) 56 Many novel forms of entertainment are given blend names. Octopush, for instance, is ‘a game similar to ice hockey in which a weight is pushed along on the bottom of a swimming pool by two rival teams of divers’: (13) He needed a sport that could be played in a pool; his idea, a game called “Octopush,” ‘octopus + push’ began to spread. (NOW, 2013) 57 and glamping is a new form of ‘camping that involves accommodation and facilities more luxurious than those associated with traditional camping’: (14) Let’s talk about glamping ‘glamorous + camping’. I know Beyonce and Jay-Z do it. (COCA, 2016) 58 One can also appreciate the ‘new form of animated films produced using the graphics engine from a video game’ (machinima): (15) […] their storyboarding, recorded video in Second Life in a machinima ‘machine + cinema’ format, and constructed three-dimensional “emotional spaces” called Storyworlds. (COCA, 2009) 59 or enjoy ‘episodes of a comedy series, which are made available online’ (webisodes): (16) Nimbus became one of the early experimenters with the short online film, or “webisode” ‘web + episode’. (COCA, 2012) 60 Tourism ‘in which travellers spend time doing voluntary work’ is now called voluntourism: (17) Still, so-called voluntourism ‘volunteer + tourism’ is a tricky business. (COCA, 2013) 61 and ‘a person who follows a primarily but not strictly vegetarian diet’ is a flexitarian: (18) If these facts have persuaded you to eat less meat, and to choose pastured or grass-fed when you do, then you’re on the brink of becoming a flexitarian ‘flexible + vegetarian’. (COCA, 2014) Lexis, 14 | 2019 67 62 while ‘a person who eats discarded food, typically collected from the refuse of shops or restaurants’ is a freegan: (19) Surplus vegetables collected by the freegan ‘free + vegan’ group were redistributed to soup kitchens and community fridges. (NOW, 2018) 63 Fashion is another setting where new blend words may be coined, especially to find names for garments which have a hybrid appearance, such as ‘a two-piece swimsuit for women, consisting of a bikini bottom and a top part in the style of a tank top’ (tankini): (20) A bold bronze tankini ‘tank top + bikini’ inspires body confidence and attitude. (COCA, 2010) 64 or ‘a pair of shorts having a flap across the front to give the appearance of a skirt’ (skort): (21) I’m not a fashionista, as my current attire showed: denim skort ‘skirt + short(s)’, sleeveless white shirt, casual sandals. (COCA, 2013) 65 Finally, hybrid animals, cross-breeds from two species, are given iconic names, such as zonkey ‘the offspring of a zebra and a donkey’ (cf. zedonk, zebrule), beefalo ‘a cross-bred livestock animal that is three-eighths bison and five-eighths domestic cow’ (cf. catalo), or labradoodle ‘a dog cross-bred from a Labrador retriever and a poodle’: (22) The two pen pals producing as zonkey ‘zebra + donkey’, a cross between a zebra and a donkey… (COCA, 2013) (23) The Fort served perfectly medium-rare beefalo ‘beef + buffalo’. (COCA, 2004) (24) Only Sylvia’s not a woman – she’s a dog; a labradoodle ‘Labrador + poodle’, to be precise. (COCA, 2017): 66 The function of defining a new category of people (e.g. screenager or kidult) is less common, and even rarer by using a blend adjective (computerate): (25) A screenager’s ‘screen + teenager’ constant companion is the television screen, which can be used to watch the news or a cartoon… (COCA, 1997) (26) Many kidult ‘kid + adult’ use their savings to buy and display toys and figurines that used to be their childhood fascination. (NOW, 2018) (27) Najib revealed that computerate ‘computer + literate’ corporate ownership was merely 10% and yet to achieve a target of at least 30%. (NOW, 2013) 67 New blend names are also given to towers (Skylon) and cities (Queuetopia, humorously alluding to Great Britain’s queuing habits): (28) Gazing out over the falls from the Skylon ‘sky + pylon’ Tower, travellers can ride the elevator to the observation deck and enjoy a view over American Falls, Bridal Veil Falls, and Horseshoe Falls. (NOW, 2018) (29) In Queuetopia ‘queue + utopia’ – Britain – people are practically crazy for queuing. (NOW, 2017) 68 In product and brand names, we find general instances, such as: Lexis, 14 | 2019 68 (30) I got a koozie ‘cool + cosy’, a beer koozie, that says storm on it. (COCA, 2012) (31) Amazon is reportedly developing a freemium ‘free + premium’ version of Prime Video. (COCA, 2017) 69 But we also find trademarks of, especially, foods and drinks: (32) Their MeritageTM ‘merit + heritage’ Espresso is the inspiration for Rashelle’s creamy mousse. (COCA, 2011) (33) […] a Pinot Noir with a Hermitage vine to develop the PinotageTM ‘Pinot + hermitage’, South Africa’s signature grape. (COCA, 2012) (34) Add gin or vodka, dash of Tabasco, Worcestershire, ClamatoTM ‘clam + tomato’ juice and horseradish (optional). (NOW, 2018) 70 or other types of product: (35) The ThinsulateTM ‘thin + insulate’ pants could not be doing much good against the cold, because they left very little to the imagination. (COCA, 2016) (36) “PsychedelicatessenTM” ‘psychedelic + delicatessen’ – Advance tickets ($ 12) for the Jan. 12 rave dance at International Ballroom are available at Ticketmaster. (COCA, 1992) (37) He wrote his first three books on an IBM SelectricTM ‘select + electric’ typewriter, which was more his style. (COCA, 2010) 71 Thus confirming the predominant role of blending for naming/labelling purposes, as anticipated in § 2. 3.3. Blends and analogy 72 From the formal viewpoint, some blends can be accommodated within the model of analogy in word-formation elaborated in Mattiello [2017]. In particular, this sub- section is devoted to the distinction between blends that are analogical to a precise model word – sharing formal (morphotactic) and semantic similarity with it – and blends that are instead created after a series of words which act as schema model. The former type is obtained by surface analogy (after Motsch’s [1981: 101] “Oberflächenanalogie”) and the latter is created via a schema (cf. Köpcke [1993] for “schema” in inflectional morphology). A schema does not have the same level of abstraction as word-formation rules, but consists of two or more words which function as concrete prototypes for novel formations. 3.3.1. Surface analogy 73 Bauer [1983: 96] defines an analogical formation as “a new formation clearly modelled on one existing lexeme”. Compounds such as whitelist, coined after its antonym blacklist, well illustrate this phenomenon. However, like compounds, new blends can also be created on a precise model. Consider, for instance, the blends smaze (← smoke + haze) and vog (← volcanic + fog), which have been coined after the exact model word smog (← smoke + fog), lexicalized in English. The similarity relation between model (smog) and targets (smaze, vog) can be analyzed as a paradigmatic substitution in the equations: smoke ˄ fog : smog = smoke ˄ haze : X (X = smaze) Lexis, 14 | 2019 69 smoke ˄ fog : smog = volcanic ˄ fog : X (X = vog) 74 In these analogical proportions, the paradigmatic substitution of fog with haze gives the blend smaze, while the replacement of smoke with volcanic gives vog. Both new blends share with their model 1) a formal resemblance, merging a word beginning with a word end, 2) a phonological resemblance (onset sm- /sm/ in the first blend and rhyme -og / ɒɡ/ in the second blend), and a semantic similarity, in that, like smog, which iconically refers to ‘fog intensified by smoke’, smaze is ‘a mixture of smoke and haze’ and vog is ‘fog containing volcanic dust’. However, while smaze is, like smog, a coordinate blend combining two nouns, vog is a headed blend, with an adjective (volcanic) modifying the head fog. From the diachronic viewpoint, the targets smaze [1953] and vog [1969] follow the model smog [1905]. 75 Other novel blends in our database which are coined by surface analogy include: • blaxploitation [1972] ‘the exploitation of black people’ ← black + exploitation, after sexploitation [1924] ‘sexual exploitation’ ← sex + exploitation; • Britcom [1977] ‘a comedy film produced in the United Kingdom’ ← sitcom [1964] ‘situation comedy’, here the model is a clipped compound (cf. the analysis of Britcom as a blend from British and sitcom; see also romcom [1971] ‘romantic comedy’); • Motopia [1959] ‘an urban environment designed to meet the needs of a pedestrian society by strict limitation of the use of the motor car’ ← motor + utopia, after Subtopia [1955] ‘Suburbia regarded as an undesirable or unattractive place to live’ ← suburb + utopia; • politicide [1967] ‘the killing of a particular group because of its political beliefs’ ← political + homicide, after genocide [1944] ‘the deliberate and systematic extermination of an ethnic or national group’ ← genus + homicide; • ragazine [1987] ‘a magazine of inferior quality’ ← colloquial rag + magazine, after fanzine [1949] ‘a magazine for fans’ ← fanatic + magazine; • slumpflation [1974] ← slump + inflation, after stagflation [1965] ← stagnation + inflation (see § 3.2.2.); • threequel [1983] ‘the third of a sequence of films’ ← three + sequel, after prequel [1958] ‘a book, film, etc., narrating events which precede those of an already existing work’ ← pre- + sequel (with a substitution of the prefix pre- with the rhyming numeral three). 76 These examples show how analogy increases regularity and helps predictability in blend formation. The type of blend obtained after a schema model is even more regular and productive, in that it creates novel splinters. 3.3.2. Analogy via schema, series, and splinters 77 While Bauer [1983: 96] claims that analogy does not give rise to productive series and Plag [1999: 210] argues that “analogical formations should be distinguished from instantiations of productive word formation rules”, analogy via schema can originate series, i.e. formations which share the same process. In particular, analogy can produce novel “splinters”, defined by Bauer et al. [2013: 525] as “non-morphemic portions of a word that have been split off and used in the formation of new words with a specific new meaning”. Splinters can be obtained by mere abbreviation of a word, as in -ware (← software), used in the creation of freeware, shareware ‘software which is available free of charge’, and vapourware ‘piece of software which, despite being publicized or marketed, does not exist’, in which the meaning conveyed is that of the full form ‘software’. Another way to obtain splinters is by secretion, i.e. reinterpretation of a word part. For Lexis, 14 | 2019 70 instance, the splinter -gram (← telegram) has extended from the meaning ‘a message sent by telegraph’ to ‘a message delivered by a representative of a commercial greetings company, esp. one outrageously dressed to amuse or embarrass the recipient’, as in kissogram ‘a greetings message delivered with a kiss’ or strippergram ‘message delivered by a performer of strip-tease’. Productive splinters in our database include: • -bot (← robot) ‘automated program which searches out information’, as in infobot [1986] ‘any of various automated systems for obtaining information’, knowbot [1988] ‘program designed to search and retrieve information from the Internet’, cancelbot [1993] ‘a program that searches for and deletes specified postings from Internet newsgroups’, and the nonce words searchbot, googlebot, etc. found in the OED. • -ercise (← exercise) ‘physical or non-physical but strenuous activity’, as in sexercise [1942] ‘sexual activity regarded as exercise’, dancercise [1967] ‘dancing performed as an exercise’, followed by jazzercise [1976] ‘a programme of physical exercises designed to be carried out in a class to the accompaniment of jazz music’ ← jazz + -ercise, and boxercise [1985] ‘a form of aerobic fitness routine incorporating exercises from boxing training’ ← box + -ercise. • -kini (← the atoll of Bikini, reanalysed as having a prefix bi-) ‘type of swimsuit or beach garment for women’, as in monokini [1964] ‘a one-piece beach garment or swimming costume worn usually by women’, trikini [1967] ‘any of various designs of ladies’ swimsuit which consist of three main areas of fabric’, and analogical tankini [1985] ← tank top + -kini (see § 3.2.2.) and burkini [2002] ‘a type of swimsuit for women which covers the head and body’ ← burka + -kini. • -lish (← English) ‘variety of English displaying features of other languages’, as in Spanglish [1933] ‘a mixture of Spanish and English’, originating Chinglish [1957] ‘a mixture of Chinese and English’ ← Chinese + -lish, Japlish [1960] ‘English language spoken in an unidiomatic way by a Japanese speaker’ ← Japanese + -lish, Hinglish [1967] ‘a mixture of Hindi and English’ ← Hindi + -lish, Singlish [1984] ‘an informal variety of English spoken in Sri Lanka, incorporating elements of Sinhala’ ← Sinhalese + -lish, or Singlish [1984] ‘an informal variety of English spoken in Singapore’ ← Singaporean + -lish. • -(o)nomics (← economics) ‘the economic policies of a President or head of state’, as in Nixonomics [1969] ‘the economic policies of Richard Nixon’, and analogical Reaganomics [1970] ← (Ronald) Reagan + -nomics, Clintonomics [1992] ← Clinton + -nomics, Rogernomics ← Roger (Owen Douglas, New Zealand Minister of Finance) + -nomics. This splinter is often blended with words that end in n (Nixon, Reagan, Clinton), thus creating overlap blends. • -tainment (← entertainment) ‘genre of broadcasting in which entertainment is combined with another genre’, as in docutainment [1978] ‘a film which includes documentary materials, and seeks both to inform and to entertain’, infotainment [1980] ‘broadcast material which seeks to inform and entertain simultaneously’, edutainment [1983] ‘informative entertainment’, and analogical eatertainment [1992] ‘an experience which combines eating with entertainment’ ← eat + -(er)tainment, irritainment [1993] ‘broadcast material which is irritating yet still entertaining’ ← irritating + -tainment, and shoppertainment [1993] ‘the provision of entertainment within a shopping centre’ ← shopper + -tainment. • -tarian (← vegetarian) ‘someone with a diet restriction’, as in dietarian [1880] ‘one who lives in accordance with prescribed rules for diet’, fruitarian [1893] ‘one who lives on fruit’, nutarian [1909] ‘vegetarian whose diet is based on nut products’, and more recent breatharian [1979] ‘a person who consumes no nutrients other than those absorbed from the air’ ← breath + -arian and flexitarian [1998] ‘a person who follows a primarily but not strictly vegetarian diet’ ← flexible + -tarian. Lexis, 14 | 2019 71 • -tel (← hotel) ‘accommodation which functions as a hotel’, as in motel [1925] ‘a roadside hotel catering primarily for motorists’ and boatel [1950] ‘a ship or boat which functions as a hotel’, acting as models for floatel [1959] ‘a floating hotel’ ← float + -tel and apartotel [1965] ‘a type of hotel which offers private suites for self-catering’ ← apartment + -tel. • -umentary (← documentary) ‘programme which has the characteristics of a documentary but also of another genre/subject’, as in mockumentary [1965] ‘a programme which adopts the form of a documentary in order to satirize its subject’, and later rockumentary [1969] ‘a documentary film on the subject of rock music’ ← rock + -umentary, and shockumentary [1970] ‘a documentary film with shocking subject’ ← shock + -umentary. 78 Examples of these formations are also attested in corpora. For example, in the NOW corpus we find additional instances, mainly nonce words or hapax legomena, which display the same splinters: (38) According to Adelana, the chatbot contributes to facilitating and accelerating the process of providing information to market participants. (NOW, 2018) (39) If you find you are little demotivated at work, do some deskercise or even go out for a brisk walk at lunch time. (NOW, 2018) (40) Just like wearing a mankini wouldn’t be appropriate in the office, there are some things that just aren’t acceptable in and around your neighbourhood. (NOW, 2018) (41) The policy may be translated to Taglish (Tagalog-English) or the local dialect, depending on the educational attainment of the majority. (NOW, 2018) (42) Pushing China to change that model is a key goal of Trumponomics. (NOW, 2018) (43) There is infotainment and advertainment and edutainment. Food, sports and retail have all embraced its possibilities. (NOW, 2018) (44) High-tea options include vegan, gluten-free vegetarian, pescetarian, and dairy-free. (NOW, 2018) (45) Other summer documentaries include “A LEGO Brickumentary,” the story of a toy that became a subculture. (NOW, 2015) 79 Some of these splinters are so regular that they are labelled “combining forms” in the OED (-bot, -tainment) [cf. Bauer et al. 2013]. The splinter -ercise has also become productive in English, so much so that Baldi & Dawar [2000: 968] have assigned it the label of “unconventional suffix”. In their work, they also cite creative formations such as commutercize [n.d.] and computercize [n.d.] as novel derived words. Moreover, Adams [1973: 170] describes -tarian as a “suffix” that occurs in a subgroup of words “inspired by vegetarian [1842] and having to do with ‘beliefs about diet’”. He also cites meatarian [n.d.] and sea-foodetarian [n.d.] [Adams 1973: 170] among additional examples, which cannot be considered mere blends, because their meaning involves reinterpretation of vegetarian not including ‘vegetables’. 3.4. General discussion 80 The presence of frequent splinters in blend formation shows that a process of regularization is underway. This regularization does not involve the same generalization or abstraction as in rules, but it is regularization triggered by analogy. Analogy may be based on surface (i.e. phonological, morphotactic, and semantic) similarity between two words, such as smaze and smog, but also between one novel word Lexis, 14 | 2019 72 (e.g. tankini) and a series of formations which constitute a schema (bikini, monokini, trikini, etc.). Unlike rules, a schema is a concrete template for novel formations. 81 From our analysis, it results that a schema model is especially functional in attributive or headed blends, where the first component (or variable part) qualifies or specifies the head. For instance, in -tarian formations, the first component specifies the type of diet restriction, e.g., to vegetables (vegetarian), fruit (fruitarian), nuts (nutarian), meat (meatarian), fish (pescetarian), seafood (sea-foodetarian), breathed air (breatharian), or flexible (flexitarian). Also blends that are apparently coordinate, such as apartotel, dancercise, docutainment, or Japlish, whose source words (i.e. apartment + hotel, dance + exercise, documentary + entertainment, Japanese + English) seem to have equally contributed to the meaning of the blend, are actually headed, in that the blends denote ‘a type of hotel, exercise, entertainment’ or ‘a variety of English’. In other words, the right source word (or splinter) carries a greater semantic weight than the left one, in that it corresponds to the head of an equivalent compound. This greater semantic importance is linked to the process of secretion or specialization that leads splinters to develop into combining forms or secreted affixes. 82 Thus, in blending, there is often an evolution from splinter to combining form, or secreted suffix. In other words, when a splinter is frequently and productively used to coin a series of new words, it can be viewed as a regular combining form, or even as a suffix, especially if it is not only abbreviated but also involves a semantic reinterpretation. This evolution corresponds to a development from surface analogy, with a precise model word, to analogy via schema, with several prototype words which constitute a series and function as model for new formations, both neologisms and nonce words. Needless to say, not all nonce formations become lexicalized words, but the fact that they use an existing schema as model provides further stability to the pattern and may represent the first step towards productivity and rule. 4. Conclusions 83 The blending phenomenon is varied and hard to classify within morphological modules and theoretical frameworks. Many blends indeed display an extra-grammatical nature, with an unpredictable output, not transparently analyzable into existing morphemes, and source words that are difficult to recognize. The possible combinatory patterns of the source lexemes, the different portions that are retained in the resulting blend, and their semantic contribution to the overall meaning increase the number of variables and classificatory criteria for blends, thus decreasing predictability of the output given an input. 84 Unlike regular compounds, whose input is unambiguous and unequivocal, the input of blends is generally opaque, unclear, blurred, and difficult to reconstruct due to the missing word part(s). Unlike combining forms, whose regularity allows abstraction and generalization based on the secretion process, blends do not allow the same abstraction as in rules. These are some of the reasons why blends are often overlooked in morphological theories, relegated to extra-grammatical operations of word-creation, and confined in use to the fields of advertising and humorous literature. 85 However, this study has demonstrated that 1) blends are increasing and increasingly important in specialized domains, besides being used in familiar contexts; and 2) they Lexis, 14 | 2019 73 mainly have a naming/labelling function, in other words, they are necessary either to fill a conceptual and/or lexical gap, the new name being often iconic of the referent’s meaning, or to label novel products and trademarks. 86 A lexicographic investigation combined with a corpus-based analysis have shown that blending is growing as a word-formation process, with several new blends that are lexicalized, included in the OED, and attested in corpora of English with low to high frequency. A corpus-based study has supported the quantitative results, by showing the variety of contexts that are favourable to blending as a word-formation mechanism, as well as the functions that new blends can serve and the effects that they produce. 87 A data-driven approach has also shown that some new blends resemble existing ones. This suggests that blending may be triggered by analogy, although the two processes do not coincide. For instance, wargasm and ambisextrous are blends from war + orgasm and ambidextrous + sex, not “genuine analogical formations” coined after orgasm or ambidextrous [Bauer 1983: 96]. The “phonetic resemblance” that Bauer [1983: 96] mentions is just a consequence of the fact that wargasm and ambisextrous include in their form (and meaning) the words orgasm and ambidextrous. Hence, not all blends are analogical, only those which follow a concrete model are. 88 Analogical blends, indeed, are a particular type of blend displaying phonological, morphological, and semantic similarity with another word or series. For example, blaxploitation is analogically formed after the model word sexploitation because: phonologically, they share three syllables and a coda, morphologically, they are both obtained by blending two words, and semantically they refer to the exploitation of something or someone (sex/sexual material, black people) for commercial purposes. This multilevel resemblance between the two blends helps English speakers recognize the source lexemes of the new blend (black + exploitation), leads them to accept the new blend as part of their lexical stock, and encourages them to form similar blends in the likeness of the model. 89 Hence, this study has shown that analogy can provide some regularity to the blending process and increase predictability in the formation of novel blends, especially those based on a concrete schema model. For instance, novel blends such as Rogernomics, Obamanomics, and Trumponomics can be interpreted on the basis of the existing series Nixonomics, Reaganomics, Clintonomics, etc., which has given birth to the splinter - (o)nomics referring to ‘the economic policies of a President or head of state’. Frequent splinters such as -(o)nomics, -tainment, or -tarian ease source word recognizability, facilitate blend interpretation, and encourage speakers to adopt new blends in their vocabulary. Because of their frequency and productivity in the coinage of novel lexical blends, some blend splinters deserve the label of ‘combining form’. In other words, some blends are unique instances of word-creation, others are part of productive word- formation. Lexis, 14 | 2019 74 BIBLIOGRAPHY ADAMS Valerie, 1973, An Introduction to Modern English Word-formation, London: Longman. ARNDT-LAPPE Sabine & PLAG Ingo, 2013, “The role of prosodic structure in the formation of English blends”, English Language and Linguistics 17, 537-563. BALDI Philip & DAWAR Chantal, 2000, “Creative processes”, in BOOIJ Geert E., LEHMANN Christian, MUGDAN Joachim, KESSELHEIM Wolfgang & SKOPETEAS Stavros (Eds.), Morphologie-Morphology: An International Handbook of Inflection and Word-formation, Vol. 1, Berlin/New York: De Gruyter, 963-972. BAT-EL Outi, 2006, “Blends”, in BROWN Keith (Ed.), Encyclopedia of Language and Linguistics, Oxford: Elsevier, 66-70. BAUER Laurie, 1983, English Word-formation, Cambridge: Cambridge University Press. BAUER Laurie, 2001, Morphological Productivity, Cambridge: Cambridge University Press. BAUER Laurie, 2012, “Blends: Core and periphery”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin/Boston: De Gruyter, 11-22. BAUER Laurie & HUDDLESTON Rodney, 2002, “Lexical word-formation”, in HUDDLESTON Rodney & PULLUM Geoffrey K. (Eds.), The Cambridge Grammar of the English Language, Cambridge: Cambridge University Press, 1621-1723. BAUER Laurie, LIEBER Rochelle & PLAG Ingo, 2013, The Oxford Reference Guide to English Morphology, Oxford: Oxford University Press. BELIAEVA Natalia, 2014, “A study of English blends: From structure to meaning and back again”, Word Structure 7, 29-54. BLANK Andreas, 2001, “Pathways of lexicalization”, in HASPELMATH Martin, KÖNIG Ekkehard, OESTERREICHER Wulf & RAIBLE Wolgang (Eds.), Language Typology and Language Universals, Vol. 2, Berlin: De Gruyter, 1596-1608. BOOIJ Geert E., 2010, Construction Morphology, Oxford: Oxford University Press. BRINTON Laurel J. & TRAUGOTT Elizabeth C., 2005, Lexicalization and Language Change, Cambridge: Cambridge University Press. CANNON Garland, 2000, “Blending”, in BOOIJ Geert E., LEHMANN Christian, MUGDAN Joachim, KESSELHEIM Wolfgang & SKOPETEAS Stavros (Eds.), Morphologie-Morphology: An International Handbook of Inflection and Word-formation, Vol. 1, Berlin/New York: De Gruyter, 952-956. COCA: Corpus of Contemporary American English, 1990-2017, https://corpus.byu.edu/coca/ (last accessed 07/04/2019). CONNOLLY Patrick, 2013, “The innovation and adoption of English lexical blends”, JournaLIPP 2, 1-14. DRESSLER Wolfgang U. & TUMFART Barbara, 2017, “New corpus-linguistic approaches to the investigation of poetic occasionalisms: The case of Johann Nepomuk Nestroy”, in DZIUBALSKA- KOŁACZYK Katarzyna (Ed.), Yearbook of the Poznań Linguistic Meeting 3, 155-166. Lexis, 14 | 2019 75 DRESSLER Wolfgang U., 2000, “Extragrammatical vs. marginal morphology”, in DOLESCHAL Ursula & THORNTON Anna M. (Eds.), Extragrammatical and Marginal Morphology, München: Lincom Europa, 1-10. FRADIN Bernard, 2000, “Combining forms, blends and related phenomena”, in DOLESCHAL Ursula & THORNTON Anna M. (Eds.), Extragrammatical and Marginal Morphology, München: Lincom Europa, 11-59. GRIES Stefan Th., 2004, “Isn’t that fantabulous? How similarity motivates intentional morphological blends in English”, in ACHARD Michel & KEMMER Suzanne (Eds.), Language, Culture, and Mind, Stanford, CA: CSLI, 415-428. HOCKETT Charles F., 1967, “Where the tongue slips, there slip I”, in HALLE Morris (Ed.), To Honor Roman Jakobson, The Hague: Mouton, 910-936. KEMMER Suzanne, 2003, “Schemas and lexical blends”, in CUYCKENS Hubert C., BERG Thomas, DIRVEN René & PANTHER Klaus-Uwe (Eds.), Motivation in Language: From Case Grammar to Cognitive Linguistics. Studies in Honour of Günter Radden, Amsterdam/Philadelphia: Benjamins, 69-97. KÖPCKE Klaus-Michael, 1993, Schemata bei der Pluralbildung im Deutschen: Versuch einer kognitiven Morphologie, Tübingen: Gunter Narr Verlag. LEHRER Adrienne, 1996, “Identifying and interpreting blends: An experimental approach”, Cognitive Linguistics 7(4), 359-390. LEHRER Adrienne, 2007, “Blendalicious”, in MUNAT Judith (Ed.), Lexical Creativity, Texts and Contexts, Amsterdam/Philadelphia: Benjamins, 115-133. MATTIELLO Elisa, 2013, Extra-grammatical Morphology in English. Abbreviations, Blends, Reduplicatives, and Related Phenomena, Berlin/Boston: De Gruyter. MATTIELLO Elisa, 2017, Analogy in Word-formation. A Study of English Neologisms and Occasionalisms, Berlin/Boston: De Gruyter. MATTIELLO Elisa, 2018, “Paradigmatic morphology: Splinters, combining forms, and secreted affixes”, SKASE Journal of Theoretical Linguistics, 15(1), 2-22, also available http://www.skase.sk/ Volumes/JTL36/pdf_doc/01.pdf MILROY James, 1992, Linguistic Variation and Change. Oxford: Blackwell. NOW: News on the Web, 2010-2018, https://corpus.byu.edu/now/ (last accessed 01/01/2019). OED2-3: Oxford English Dictionary Online, 1989-2019, 2nd/3rd ed. Oxford: Oxford University Press, http://www.oed.com/ (last accessed 10/10/2018). PLAG Ingo, 1999, Morphological Productivity: Structural Constraints in English Derivation, Berlin/New York: De Gruyter. RENNER Vincent, 2006, Les composés coordinatifs en anglais contemporain, Lyon: Université Lumière Lyon 2 dissertation. RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L., 2012, Cross-Disciplinary Perspectives on Lexical Blending, Berlin/Boston: De Gruyter. RONNEBERGER-SIBOLD Elke, 2006, “Lexical blends: Functionally tuning the transparency of complex words”, Folia Linguistica 40(1-2), 155-181. Lexis, 14 | 2019 76 RONNEBERGER-SIBOLD Elke, 2010, “Word creation: Definition – function – typology”, in RAINER Franz, DRESSLER Wolfgang U., KASTOVSKY Dieter & LUSCHÜTZKY Hans C. (Eds.), Variation and Change in Morphology, Amsterdam/Philadelphia: Benjamins, 201-216. THORNTON Anna M., 1993, “Italian blends”, in TONELLI Livia & DRESSLER Wolfgang U. Eeds.), Natural Morphology. Perspectives for the Nineties, Padova: Unipress, 143-155. THORNTON Anna M., 2004, “Parole macedonia”, in GROSSMAN Maria & RAINER Franz (Eds.), La formazione delle parole in italiano, Tübingen: Niemeyer, 567-571. NOTES 1. The dates in brackets refer to the earliest attestation of the words in the OED. See Beliaeva [2014] for an in-depth study of the phonological and structural differences between blends and clipping compounds. 2. The remaining 53.5% blends display a corpus frequency between 0.01 and 0.09 pmw. ABSTRACTS This study investigates lexical blending from a synchronic perspective, with special focus on new blends in English. It analyses a sample of 245 English blends dated 1950-2010, from both quantitative and qualitative viewpoints, with the purpose of reassessing the importance of the blending phenomenon in terms of 1) its suitability in the coinage of new specialized vocabulary, and 2) its regularity in the creation of words containing frequent splinters. From the theoretical viewpoint, the study contributes to the issue of whether blending should be considered an extra- grammatical phenomenon of word-creation or a regular process of word-formation. It supports the claim that while some blends (e.g. glam-ma ← glamour + grandma) are unique instances, others are part of series (e.g. eatertainment, irritainment, shoppertainment, all obtained from the splinter - tainment ← entertainment), and therefore show a tendency towards greater regularity and productivity. The goal of the study is fourfold. First, it aims at identifying the contexts/registers which favour the formation of blend words, ranging from slang/colloquial registers to specialized domains. Second, this study addresses the question of whether blends are created with the intention of designating a new referent or to give a new name to an existing referent. Third, the study focuses on a particular type of blending, called ‘attributive’ or ‘headed’, which displays an endocentric relation with its head, as in rockumentary (← rock + documentary) and Clintonomics (← Clinton + economics). In particular, some attributive blends will be viewed as possible schema models for new lexical blends, with the second splinter -umentary and -(o)nomics as potential combining forms or secreted affixes for novel formations. Cette étude examine l’amalgame lexical dans une perspective synchronique. En particulier, elle analyse une collection de 245 nouveaux mots-valises en anglais, sur la période 1950-2010, d’un point de vue quantitatif et qualitatif. Cette étude a pour objectif de réévaluer l’importance de l’amalgame 1) pour inventer un nouveau lexique spécialisé, et 2) pour créer des mots qui contiennent des fractomorphèmes réguliers. D’un point de vue théorique, l’étude tente de répondre à la question suivante : l’amalgame est-il un phénomène extra-grammatical de création Lexis, 14 | 2019 77 de mot ou un procès régulier de formation de mot ? L’étude affirme que certains mots-valises sont des cas uniques (p.ex. glam-ma ← glamour + grandma), tandis que d’autres font partie de séries de mots (p.ex. eatertainment, irritainment, shoppertainment viennent du fractomorphème - tainment ← entertainment) et tendent vers la régularité et la productivité. L’étude a quatre objectifs. D’abord, elle vise à montrer les contextes/registres qui privilégient la formation des mots-valises, et qui vont du langage familier au langage de spécialité. Deuxièmement, l’étude vise à répondre à la question suivante : les mots-valises sont-ils créés pour designer de nouveaux référents ou pour donner un nouveau nom à un référent existant ? Troisièmement, l’étude se concentre sur l’amalgamation dite ‘attributive’ ou ‘endocentrique’, comme rockumentary (← rock + documentary) et Clintonomics (← Clinton + economics). Les mots-valises attributifs sont décrits comme des modèles possibles ou schémas pour les nouveaux amalgames lexicaux, dont les seconds fractomorphèmes -umentary et -(o)nomics sont des affixes secrétifs potentiels pour de nouveaux mots. INDEX Mots-clés: mots-valises attributifs, nouveaux mots, néologismes, anglais, fractomorphème, affixes secrétifs, schéma Keywords: attributive blends, new words, neologisms, English, splinters, secreted affixes, schema AUTHOR MATTIELLO ELISA University of Pisa [email protected] Lexis, 14 | 2019 78 Variable base-word positioning in English blends Aviv Schoenfeld, Evan Gary Cohen and Outi Bat‑El Introduction 1 In the vast majority of English blends [Adams 1973, Bryant 1974, Algeo 1977, Bat‑El 2006], the order of the base words is invariable. For example, smoke × fog yield smog rather than foke, and lábrador × póodle yield lábradoodle rather than póodrador (the accented vowel letter marks the nucleus of the primarily stressed syllable; we ignore secondary stress). However, there is a small but non‑negligible number of cases where the order of the base words varies, yielding two (or more) blends, i.e. blend doublets. Cases noted in the literature include absotívely ~ posolútely (< absolútely × positívely) [Algeo 1977: 60], tígon ~ líger (< tíger × líon) [Bat‑El 2006], and chévrolac ~ cádillet (< chévrolet × cádillac) [Gries 2012: 160]. In some doublets, the blends are synonymous, e.g. plúmpricot ~ ápriplum ‘plum‑apricot hybrid’ (< plum × ápricot), while in others they are not, e.g. (egg × prégnant ⇒) éggnant ‘pregnant with egg’ ~ prégegg ‘egg that counts down pregnancy’. 2 In this paper, we explore the conditions that result in variable base‑word positioning in English blends. We argue for the relevance of the difference between synonymous vs. non‑synonymous blend doublets, and endocentric vs. non‑endocentric blends. In non‑synonymous doublets, the main factor is that in endocentric blends, the base word that contributes the semantic head is right‑aligned. For example, éggnant denotes a way of being pregnant, whereas prégegg denotes a kind of egg. By contrast, in synonymous blend doublets (where the vast majority of blends are non‑endocentric), variable base‑word positioning results from at least four factors interacting: (i) segmental faithfulness – maximizing segmental similarity; (ii) Pāṇini’s law – positioning the short base word before the long one; (iii) monosyllabic integrity – keeping the monosyllabic base word in the same syllable; and (iv) syntagmatic faithfulness – matching the linear order of the base words to their order in a would‑be syntactic constituent. Lexis, 14 | 2019 79 3 In the § 1., we introduce the corpus of the present study. We then turn to the analysis (§ 2.), where we distinguish between non‑synonymous (§ 2.1.) and synonymous (§ 2.2.) blend doublets. 1. Corpus: Monosyllabic Base Blends 4 The goal of the corpus is to determine what influences variable base‑word positioning in English blends. For this, we use blends with exactly one monosyllabic base word. The rationale behind this is that the fewer segments a base word has, the more flexible its positioning potentially is. To illustrate, consider blends with the base words entreprenéur and either mom or móther. While momtreprenéur, entrepremóm and motherprenéur occur in webpages as common nouns in full sentences in non‑metalinguistic contexts, shown in (1), entremother (stress pattern unknown) does not. Thus, the monosyllabic mom has more flexible positioning than the disyllabic móther. (1) Flexible positioning of monosyllables (mom) a. Becoming a momtrepreneur can enable you to have truly fulfilling family and professional lives. b. Ever since she could remember, entrepremom Rachelle Chua-Villaceran’s biggest dream was to have her own ice cream parlor. c. For many, many years I was the most goal oriented motherpreneur you can think of. 5 Besides exactly one base word being monosyllabic, we limit our corpus in two additional ways: (i) the blend has the same number of syllables and stress pattern as the polysyllabic base word, and (ii) the monosyllabic base word is not truncated segmentally or orthographically. Blends that violate (i) include cómpucar (< compúter × car), with a different stress pattern from compúter, and plúmcot (< plum × ápricot), with a different number of syllables from ápricot. Blends that violate (ii) include áprium (< ápricot × plum), where plum is truncated segmentally and orthographically, and bláxicen (< black × Méxican), where black is truncated orthographically. Admitting such blends into the corpus would have made it more heterogenous, which might have gotten in the way of the object of study. 6 Our corpus consists of 407 pairs of base words, 366 of which (~90%) yield one blend, and 41 (~10%) yield two blends with variable base‑word positioning, bringing the blend total to 448; 101 are from the corpus of Bat‑El and Cohen [2012], and the rest (n=347) were collected specifically for this study. 306 were found by browsing the Wiktionary list of English blends from to 7 The blends in the sub‑corpus of variable base‑word positioning (n=82) have at least one webpage occurrence that meets three criteria: (i) the blend is not a proper noun with two common nouns as base words, and it occurs (ii) in a full sentence (iii) in a non‑metalinguistic context (e.g. not a dictionary or question about whether the blend is a real word). These prevent the criterion for what counts as an attested blend from being too liberal: many blends occur as proper nouns (e.g. usernames, website or company names), not in full sentences (e.g. hashtags) or in user‑submitted online dictionaries, but less meet the aforementioned three criteria. Lexis, 14 | 2019 80 8 From the aforementioned webpage occurrences, the meaning of the blends in the sub‑corpus of variable base‑word positioning in ascertained. This turns out to be relevant for base‑word positioning, as shown in §2. The blends in the sub‑corpus of invariable base‑word positioning (n=366) are not coded for meaning. 9 To illustrate how the corpus was compiled, consider three blends from the corpus of Bat‑El and Cohen [2012]: assmósis (< ass × osmósis), guésstimate (< guess × éstimate), and spórtlon (< sport × nýlon). We found no webpage occurrence of osmósass that meets the aforementioned three criteria, so assmósis is in the sub‑corpus of invariable base‑word positioning. Next, we found such webpage occurrences of guésstimate and éstiguess, where both mean ‘guess & estimate’, so they are in the sub‑corpus of synonymous blend doublets (§ 2.2.). Finally, we found such webpage occurrences of spórtlon and nýlsport, where the former means ‘athletic sock’ and the latter ‘snowmobile suit’, so they are in the sub‑corpus of non‑synonymous blend doublets (§ 2.1.). 10 The forms in our corpus are analyzable in the morphological representation of native speakers as what linguists would classify as blends, but some are morphologically ambiguous. Unlike earlier studies, which draw categorial distinctions between blends and affixed forms or clipped compounds, we recognize morphologically ambiguity. For example, Bat‑El and Cohen [2012: 67] write that their corpus does not include affixed forms, what they call COMBINING FORMS, but their corpus includes irritáinment and digitéria, which are analyzable as including the suffixes ‑táinment and ‑téria; the former has an entry in Wiktionary, and the latter has an entry in the Oxford English Dictionary, with citations as early as 1923. Morphological ambiguity ‘in action’ is reported in Danks [2003: 174], who recounts a scenario where her sister perceived sluggery as a blend of wormery and slug, whereas Danks herself perceived it as an affixed form with slug and ‑ery. Perceiving sluggery as a blend or affixed form is reducible to different morphological and mental representations. The affixed representation /slʌɡ‑ɹi/ goes along with a mental activation of slug and ‑ery, whereas the blend representation / slʌɡɹi/ goes along with a mental activation of slug and wormery. Our corpus includes such morphologically ambiguous items, and in § 2.2. we explain several cases of variable base‑word positioning by appealing to word formation processes other than blending. 2. Data and Analysis 11 In this section, we present and analyze the 41 blend doublets with variable base‑word positioning. 15 include non‑synonymous blends, such as éggnant ‘pregnant with egg (bird)’ ~ prégegg ‘egg that counts down pregnancy’ (< egg × prégnant), and 26 include synonymous blends, such as plúmpricot ~ ápriplum ‘plum‑apricot hybrid’ (< plum × ápricot). In the next two subsections, we treat each sub‑corpus in turn. 2.1. Non‑synonymous blend doublets (n=15) 12 To discuss non‑synonymous blend doublets, we introduce the notion of a semantic head (see Williams [1981] for a notion of a syntactic head in compounds and affixed forms). We begin with the doublets where both blends have a semantic head (n=7), then continue to those where one or neither has a semantic head (n=8). Lexis, 14 | 2019 81 2.1.1. Both blends are endocentric 13 In 7 of the 15 non‑synonymous blend doublets (Table 1), both blends are such that their semantic head is contributed by exactly one base word (henceforth ENDOCENTRIC BLENDS), and this semantic head is aligned to the right edge of the blend. MonoL and MonoR in the column headers indicate blends where the monosyllabic base word is aligned to left and right edge of the blend respectively (apologies for offensive language in Table 1 and throughout). Table 1: Non‑synonymous blend doublets – Semantic head right‑aligned The 7 blend doublets in Table 1 indicate the activity of the constraint formulated in (2), which captures the language‑specific fact about English that soap dish denotes a kind of dish, whereas dish soap denotes a kind of soap. This constraint is formulated generally enough to apply to both blend and compounds. (2) HEADR: every base word that corresponds to a semantic head is aligned to the right edge of the form (compound or blend). 14 We conceive of (2) as a violable constraint (in the spirit of Optimality Theory [Prince and Smolensky 1994/2004]), rather than principle, because it is violable in blends [Shaw 2013: 58]. For example, a so‑called LEFT‑HEADED BLEND in Shaw’s corpus is entremanúre (< entreprenéur × manúre), a type of entrepreneur rather than manure. Our corpus includes four left‑headed blends (Table 2). Table 2: Semantic head left‑aligned Two notes are in order on the blends in Table 2. First, cánacunt and connécticunt from Table 1 reoccur in Table 2, but with different meanings. Cánacunt occurs (in webpages) as meaning either ‘Canadian cunt’ or ‘Canada, derogatory’, and only the former obeys HEADR. Second, of the four blends, three have at least one proper noun base word: cánacunt and connécticunt have one, and obámbush has two. It is unclear whether this is indicative of a larger pattern; we leave this for future research (see Moreton et al. [2017] for proper nouns in blends). At any rate, tótstitute is the odd one out, because neither base word is a proper noun. Lexis, 14 | 2019 82 2.1.1. One or neither blend is endocentric 15 Because HEADR is formulated as universal quantification over base words, it does not influence base‑word positioning in non‑endocentric blends, where both or neither base words contribute a semantic head. When both base words contribute a semantic head, as in COORDINATIVE blends like jewfugée ~ refujéw ‘Jew & refugee’ (< jew × refugée), the initial base word in both blend alternatives makes the blend violate HEADR once (see Renner [2008] for coordinative compounds). Likewise, when neither base word contributes a semantic head, as in EXOCENTRIC blends like átpersand ~ ámpersand ‘the @ symbol’ (< at × ámpersand), both blend alternatives vacuously obey HEADR. Given the identical effect of HEADR in coordinative and exocentric blends, the distinction between them is irrelevant for present purposes, so we combine them under the title NON‑ENDOCENTRIC blends. This relieves us from needing to determine for each non‑endocentric blend whether it is coordinative or exocentric, which is not always straightforward. For example, conceiving of bárstaurant (< bar × réstaurant) as both a bar and restaurant is a coordinative conception, but conceiving of it as a third separate category is an exocentric conception. 16 In the 7 doublets in Table 1, both blends are endocentric. However, in the 8 other doublets in the sub‑corpus of non‑synonymous doublets (Table 3), one (a‑f) or neither (g‑h) blend is endocentric, and those that are endocentric (shaded) obey HEADR. Table 3: One or neither blend in doublet is endocentric To reiterate, in the non‑endocentric blends (not shaded) in Table 3, HEADR does not influence base‑word positioning. Consider a blend that denotes a brand of athletic socks with the base words nylon and sport. An athletic sock is not a kind of nylon nor sport, so spórtlon and nýlsport both satisfy HEADR vacuously. In practice, spórtlon was chosen, for reasons that are irrelevant for the present study. However, what is relevant is the reason that blend doublets like spórtlon and nýlsport are attested. In cases where HEADR has no effect, it is better to examine synonymous blend doublets, because something about the meanings of ‘athletic sock’ and ‘snowmobile suit’ might support spórtlon and n ýlsport respectively. For this, we turn to synonymous blend doublets in the next subsection. 2.2. Synonymous blend doublets (n=26) 17 In this subsection, the synonymous blend doublets are analyzed in three chunks, distinguished by the degree of segmental similarity between the monosyllabic and polysyllabic base words, where similarity is in terms of a proportional string edit Lexis, 14 | 2019 83 distance. First, we treat the cases where the monosyllabic base word is most segmentally similar to a non‑initial part of the polysyllabic base word (Table 5). Next, we treat those where the monosyllabic base word is more or less equally segmentally (dis)similar to two parts of the polysyllabic base word (Table 6). Finally, we treat those where the monosyllabic base word is most segmentally similar to the initial part of the polysyllabic base word (Table 7). 18 In the sub‑corpus of synonymous blend doublets (n=26), the vast majority of blends are non‑endocentric (all but tótstitute ~ próstitot), meaning HEADR does not affect base‑word positioning. In what follows, we introduce two constraints that are known to be active in blends – segmental faithfulness and Pāṇini’s law (short before long) – and show the circumstances under which their interaction yields variable base‑word positioning. Afterwards, we introduce two constraints that are apparently strong enough to override both these constraints: monosyllabic integrity and syntagmatic faithfulness. 19 Beginning with segmental faithfulness, recall from § 1. that our corpus is constrained as follows: (i) exactly one base word is monosyllabic, (ii) the blend has the same number of syllables and stress pattern as the polysyllabic base word, and (iii) the monosyllabic base word is not truncated segmentally or orthographically. This means that not all segments of the polysyllabic base word are preserved in the blend, with the exception of blends that are homophonous with their polysyllabic base word, such as arkeólogist (< archeólogist × ark). Put differently, the vast majority of these blends are segmentally unfaithful to their polysyllabic base word. There are three such ways to be unfaithful: segment‑zero mismatches, segment‑segment mismatches, and linearity mismatches. In what follows, we illustrate these mismatches, and explain how they enter into calculating a PROPORTIONAL STRING EDIT DISTANCE (PSED), which quantifies dissimilarity between strings [Levenshtein 1966]. 20 To illustrate segment‑zero mismatches (Figure 1), consider two blends from the sub- corpus of invariable base-word positioning: barcáde (< arcáde × bar) and eýeborg (< cýborg × eye); each has a segment‑zero mismatch between it and its polysyllabic base word. In barcáde, the zero is in the polysyllabic base word, whereas in eýeborg it is in the blend. In the mismatch illustrations, the first line is the polysyllabic base word, the second line is the blend, and the monosyllabic base word is underlined. Figure 1: Two segment‑zero mismatches In the PSED (recounted fully in the appendix), segment‑zero mismatches cost 1, and the proportional edit distance between two strings is the cost divided by the number of 1 segments in the longer string. Thus, the PSED between árcade and barcáde is ⁄7=0.14, and the same goes for cýborg and eýeborg. The PSED runs between 0 and 1, where 0 is identity and 1 is complete dissimilarity. The PSED is non-directional, so it does not distinguish between zero‑segment (árcade – barcáde) and segment-zero (cýborg – eýeborg) mismatches. See Bailey and Hahn [2005: 346] for a discussion on symmetric vs. asymmetric similarity measures. Lexis, 14 | 2019 84 21 Next, to illustrate segment‑segment mismatches (Figure 2), consider bluenétte (< brunétte × blue) and búycott (< bóycott × buy); each has a segment‑segment mismatch between the blend and its polysyllabic base word – consonantal in bluenétte, and vocalic in boycott. Figure 2: Two segment‑segment mismatches In the PSED, segment‑segment mismatches cost between 0‑1, depending on the number of differing distinctive features between the segments. We use Bailey and Hahn’s [2005] four‑feature system for consonants (Appendix Tables 10‑11), and a five-feature system for vowels (Appendix Tables 12‑13). Per these systems, [ɹ‑l] mismatches cost 0.2, 0.2 meaning the PSED between brunétte and bluenétte is ⁄6=0.03, and [ɔ‑a] mismatches 0.5 cost 0.5, meaning the PSED between bóycott and búycott is ⁄6=0.08. 22 Lastly, entrepornéur (< entreprenéur × porn) has a rare case of a linearity mismatch (Figure 3), as well as a base word embedded in the middle of a blend. Our corpus of 448 blends has 7 with an embedded base word, the others being margabéerta, adórkable, edjewcátion, ambiséxtrous, entertóynment and induhvídual (the monosyllabic base word is underlined). Figure 3: Two linearity mismatches, one segment‑segment mismatch In the PSED, a linearity mismatch costs 1. The above blend also has the segment‑segment mismatch [ə‑ɔ], which costs 0.17, so the PSED between entreprenéur 1.17 and entrepornéur is ⁄10=0.12. 23 Segmental similarity affects base‑word positioning by favoring blends that maximize segmental similarity between them and their polysyllabic base word, as stated in (3) (for other versions, see Bat‑El [1996: § 5.1.], Piñeros [2004] and Gries [2004: § 3., 2012: § 4.1.]): (3) FAITHPOLY: maximize segmental similarity between blend and polysyllabic base word (measured by proportional string edit distance). 24 FAITHPOLY is a combination of several constraints, which together have the effect of similarity maximization. There is the requirement for input-output identity, which means segmental identity between the base words and blend. It is, however, impossible to achieve total identity while preserving all segments from the two base words and restricting the size of the blend to that of the polysyllabic base word. Therefore, identity is maximized but rarely achieved (except cases like the aforementioned arkeólogist < archeólogist × ark). 25 FAITHPOLY, at least partially, explains why the alternatives to the four blends in Figures 1‑2 are unattested and intuitively sound bad: *arbár (instead of barcáde), *cýeye Lexis, 14 | 2019 85 (instead of eýeborg), *brublúe (instead of bluenétte) and *bóybuy (instead of búycott). FAITHPOLY might just be a partial explanation, because something additional might be wrong with these unattested blends, e.g. something phonotactically wrong with cýeye [sajaj]. At any rate, Table 4 shows that the four blends in Figures 1‑2 are more segmentally similar to their polysyllabic base word than their unattested alternative (green indicates closer to 0, i.e. identity, and red indicates closer to 1, i.e. complete dissimilarity): Table 4: FAITHPOLY illustration Just as segmental similarity between a blend and its polysyllabic base word is a factor, so might be similarity between a blend and its monosyllabic base word. In our corpus, the PSEDs in this regard differ when the blend alternatives have a different number of 6 segments. For example, the PSED between plúmpricot and plum is ⁄10=0.6, whereas that 4 between ápriplum and plum is ⁄8=0.5. Thus, a FAITHMONO constraint can be formulated, parallel to FAITHPOLY in , but we have not found it necessary to appeal to FAITHMONO to explain variable base‑word positioning in our corpus. 26 The second constraint that is known to be active in blends is Pāṇini’s law, developed originally for coordinative compounds by the ancient Sanskrit grammarian Pāṇini (circa 350 B.C.). Cooper & Ross [1975: 78] state the law as follows: “other factors being nearly equal, place 1 elements contain fewer syllables than place 2 elements.” In their study of conjoined elements (aka binomials) in English (e.g. kit and caboodle, *caboodle and kit), Pāṇini’s law has been found to be the strongest phonological principle to determine ordering, overriding six other principles [Cooper & Ross 1975: 79]. Pāṇini’s law has also been found to be active in English blends by Kelly [1998: 582], Gries [2004: 421] and Renner [2014: § 3.2.1.]. This law manifested in our corpus as the constraint stated in (4), a specific case of Pāṇini’s law. (4) MONOL: the monosyllabic base word is aligned to the left edge of the blend. 27 In our corpus, there are two manifestations of MONOL. First, in the sub‑corpus of invariable base‑word positioning (n=366), the monosyllabic base word is left‑aligned in 288 (79%) of the blends. Note that this sub‑corpus includes endocentric blends like átrogene ‘gene that influences muscle atrophy’ (< átrophy × gene), where the monosyllabic base word gene is right‑aligned for semantic reasons, namely to respect HEADR. If such blends were not considered, the percent of blends that respect MONOL might be higher (recall that the blends in this sub-corpus are not coded for meaning). 28 The second manifestation of MONOL in our corpus comes from the seven doublets in the sub‑corpus of synonymous blend doublets (n=26), where the monosyllabic base word is most segmentally similar to a non‑initial part of the polysyllabic base word (Table 5); in (f) a middle part, and a final part in the rest. In such cases, MONOL directly conflicts with FAITHPOLY: the former pressures the monosyllabic base word to be initial, while the latter pressures it to be non‑initial. The variable base‑word positioning in Table 5 is Lexis, 14 | 2019 86 attributed to this conflict. The number to the right of each blend is the PSED between it and its polysyllabic base word, and Δ is the difference between the two PSEDs, which is greater than 0.05. Table 5: MONOL and FAITHPOLY conflict Not all cases of variable base‑word positioning in the sub‑corpus of synonymous blend doublets can be attributed to a conflict between FAITHPOLY and MONOL, as in Table 5. Consider the eight doublets in Table 6, where the two PSEDs are more or less equal (the arbitrary cutoff point is 0.05 or less). In such cases, FAITHPOLY has no effect, because both blend alternatives yield around the same PSED. Evidently, MONOL is not the sole determining factor, because it loses in the blends in the “MONOL loses” column. In what follows, we discuss what other factors might be at play, causing MONOL to lose. Table 6: Why does MONOL lose? At this point, it is fruitful to explicate the status of Pāṇini’s law (and MONOL by extension), specifically what it means to violate it. Cooper & Ross [1975] found its activity in binomials such as kit and caboodle vs. *caboodle and kit. But what is the status of the latter, i.e. what does the asterisk mean? Recall that our criteria for entering the sub‑corpus of variable base‑word positioning is relatively liberal: the blend must have at least one webpage occurrence where (i) it is not a proper noun with two common nouns as base words, and it occurs (ii) in a full sentence (iii) in a non‑metalinguistic context. By that criteria, both kit and caboodle and caboodle and kit are attested, because the latter is attested in “The whole caboodle and kit stopped in the middle of the road for a minute or two while nothing much else happened.” [Marsh, 2002: 230]. In light of that, blends that violate MONOL are not ungrammatical. Rather, they are predicted to intuitively sound worse than those that obey MONOL, all other things being equal. Thus, ápriplum is predicted to sound worse than plúmpricot, if nothing else were at play besides MONOL. Lexis, 14 | 2019 87 29 To check what other factors might be at play besides MONOL, we subjected the 8 blend doublets in Table 6 to a blend preference experiment. Participants were given definitions reworded from Table 6 (e.g. a bar and restaurant), and had to choose from between two blends which one sounded better to them (e.g. barstaurant or restaurbar). In all definitions, the monosyllabic base word occurred before the polysyllabic one, and in all blend options the MonoL blend occurred first. 30 self‑reported native monolingual English speakers took part in the experiment, and the results are shown in Table 7. Table 7: Blend preference experiment results (30 participants) Three of the eight results in Table 7, specifically (b‑d), show an 80% or so preference for the MonoL blend. Thus, we stipulate that MONOL is the sole determining factor in these cases from among the constraints considered here. The task is to explain the five results that deviate from 80%‑20%. 30 When the preference for MonoL exceeds 80%, as in (a), there are two strategies to explain this deviance from 80%‑20%: (i) the MonoL blend has an additional advantage (other than satisfying MONOL), or (ii) the MonoElse blend has an additional disadvantage (other than violating MONOL). For (a), we opt for strategy (i), and suggest that the increased preference for bushbáma is influenced by the correlation between temporal and linear order (President Bush came before President Obama). 31 Similarly, when the preference for MonoElse exceeds 20%, as in (e‑h), there are two strategies to explain this deviance from 80%‑20%: (i) the MonoL blend has a disadvantage that detracts from it obeying MONOL, or (ii) the MonoElse blend has an advantage that compensates for it violating MONOL. Thus, the increased preference (above 20%) for the MonoR réstaurbar in (e) might be influenced by the existing term restobar (strategy ii). As for the increased preference for the MonoR ámpersat in (f), perhaps this is due to dispreferring átpersand, because the [t] at the end of [æt] is unaspirated, making this base word harder to recognize (strategy i). Next, the increased preference for the MonoR philósofool in (g) might be because the monosyllabic base word occupies its own syllable, unlike in foolósopher, where the string [ful] is split across two syllables (strategy i). Finally, the increased preference for cósmodog in (h) is likely due to it being analyzable not just as a blend, but also as a prefixed form including cosmo‑, which must be initial. 32 In conclusion, when FAITHPOLY has no effect (because both blend alternatives are more or less equally (dis)similar to the polysyllabic base word), variable base‑word positioning can arise from MONOL interacting with several other factors. Of these, we argue that MONOSYLLABIC INTEGRITY explains further data in our corpus, reviewed next. Lexis, 14 | 2019 88 33 The third major group of blends in the sub‑corpus of variable base‑word positioning (n=26) are the eight doublets in Table 8, where the monosyllabic base word is most similar to the initial part of the polysyllabic base word. Thus, FAITHPOLY and MONOL both influence the monosyllabic base word to be left‑aligned, but it is also attested as right‑aligned. Table 8: Why do MONOL and FAITHPOLY lose? Two notes on Table 8 are in order. First, the PSED difference in (f) is less than 0.05, so this doublet technically belongs in Table 6. We nevertheless include it here, because [ɻɑk] rock has an overlapping vowel with the initial part of [ɑltɻ̩nətɪv] alternative. Second, we gloss (e‑f) as coordinative, e.g. (f) soultérnative and altérnasoul denote a hybrid of soul and alternative music. However, the MonoR blends in (e‑f) might turn out to be endocentric, in which case altérnasoul would denote alternative soul music, and (e‑f) would belong in the sub‑corpus of non‑synonymous doublets. Lacking the musical expertise to distinguish between a hybrid of alternative and soul music and alternative soul music, we gloss the blends in (e-f) as coordinative for simplicity’s sake. 34 The attestedness of the MonoR blends in Table 8 indicates which factors are apparently strong enough to override both FAITHPOLY and MONOL, which in these cases conspire to prefer the MonoL blend. In what follows, we suggest two such factors, the first being (5) monosyllabic integrity, which has already been mentioned with respect to (f) in Table 6. (5) MONOINTEG: The edges of the monosyllabic base word correspond to the edges of the same syllable. 35 Previously, we suggested that MONOINTEG influenced participants in our experiment to exhibit a preference for philósofool that is greater than 20% (53%). In Table 8, MONOINTEG can explain the occurrence of (a) ámbucab and (b) cárdicoat, as the MonoL alternatives (cábulance and cóatigan) violate MONOINTEG. 36 The second constraint that we suggest is responsible for the existence of the MonoR blends in Table 8 is formulated in (6), where SYNT stands for syntagmatic. (6) SYNT: if the base words can form a syntactic constituent, then their order in the blend mirrors their order in that constituent. 37 The doublets in Table 8 for which (6) is relevant are (c‑d) and (f‑g), with consérvative or altérnative as a base word. In these cases, the base words can form a syntactic constituent where the monosyllabic base word is right‑aligned (e.g. conservative cuck, alternative rock), so only the MonoR blends obey SYNT. Lexis, 14 | 2019 89 38 The two MonoR blends in Table 8 that we have yet to account for are (e) éstiguess and (h) ádverspam. As for the former, perhaps it is appealing enough to override both FAITHPOLY and MONOL because of its internal rhyme. Presently, we have no explanation for the existence of ádverspam. 39 So far, we have treated 23 of the 26 doublets in the sub‑corpus of synonymous blend doublets. This leaves the three doublets in Table 9, with éxercise as a base word: Table 9: Doublets with éxercise We propose to account for the variable base‑word positioning in Table 9 with the idea that exercise has yielded two SECRETED AFFIXES [Haspelmath 1995: 15, Fradin 2001: 46, Schoenfeld forthcoming]: the prefix éxer‑ and suffix ‑ercise. Secreted affixes (per the definition in Schoenfeld [forthcoming]) are affixes that exist in a language alongside a content word that has participated in their diachronic derivation, called the MODEL WORD. Secreted affixes are also known as COMBINING FORMS [Warren 1990, Bat‑El 1996: 317, Tomaszewicz 2008, Miller 2014: § 13.], PRODUCTIVE SPLINTERS [Lehrer 1996, 2007], SPLINTER‑ORIGINATING AFFIXES [Danks 2003] and RECURRING SPLINTERS [Bauer 2006]. Leading examples include ‑holic (model word alcoholic), as in foodoholic ‘person addicted to food’, and franken‑ (model word Frankenstein), as in frankenfood ‘genetically modified food’. 40 Rarely, a model word yields a prefix and affix. The only English example that we were aware of until now was economics, which has yielded the prefix econo‑ [1964], as in econobabble [1992] and econo‑nerds [2003] (years are from the Oxford English dictionary), as well as the suffix ‑onomics, as in bleakonomics and Trumponomics (see Schoenfeld [forthcoming] for an analysis of complex Biblical Hebrew personal names as involving prefix-suffix pairs that originate from the same model word). Parallel to economics, we suggest that exercise has yielded the prefix éxer‑ and suffix ‑ercise, which explains the variable base‑word positioning in Table 9. Conclusion 41 In this paper, we explored the conditions that result in variable base‑word positioning in English blends, an issue that has not been addressed extensively in the literature of blends. We argued that in non‑synonymous blend doublets, the main factor is that in endocentric blends, the base word that contributes the semantic head is right‑aligned (HEADR). By contrast, with synonymous blend doublets (where the vast majority of blends are non‑endocentric), variable base‑word positioning results from at least four factors interacting: (i) maximizing segmental similarity between blend and polysyllabic base word (FAITHPOLY), (ii) MONOL, a manifestation of Pāṇini’s law (short before long), (iii) keeping the monosyllabic base word in the same syllable (MONOINTEG), and (iv) faithfulness of the order of the base words to their order in a would‑be syntactic constituent (SYNT). Lexis, 14 | 2019 90 42 Variable base‑word positioning in synonymous blend doublets is unique to blends, because in other word‑formation processes, the order of the morphemes is rigid, determined mostly by lexical categories and selectional restrictions. That is, it is rare to find an affix that freely attaches to either the left or right of the base, depending on the phonological structure of the output (e.g. a CV affix, prefixed when the base begins with a consonant and suffixed when the base begins with a vowel; cf. Kim [2010]). In blends, however, phonology plays a major role in determining the order of the base words (albeit less so in endocentric blends), and phonologically-condition variation is thus expected. BIBLIOGRAPHY ADAMS Valerie, 1973, An introduction to Modern English word-formation, London: Longman. ALGEO John, 1977, “Blends, a structural and systemic view”, American speech 52, 47-64. BAT-EL Outi, 1996, “Selecting the best of the worst: the grammar of Hebrew blends”, Phonology 13.3, 283-328. BAT-EL Outi, 2006, “Blend”, in BROWN Keith (Ed.), Encyclopedia of language and linguistics, Oxford: Elsevier 2nd ed., 66-70. BAT-EL Outi & COHEN Evan-Gary, 2012, “Stress in English blends: a constraint- based analysis”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre (Eds.), Cross-disciplinary perspectives on lexical blending, Berlin: De Gruyter Mouton, 193-212. BAILEY Todd & HAHN Ulrike, 2005, “Phoneme similarity and confusability”, Journal of memory and language 52(3), 339‑362. BAUER Laurie, 1983, English word-formation, Cambridge: Cambridge university press. BAUER Laurie, 2006, “Compounds and Minor Word-formation Types”, in AARTS Bas & MCMAHON April (Eds.), The handbook of English linguistics, Malden: Blackwell, 483‑506. BRYANT, Margaret M., 1974, “Blends are increasing”, American Speech 49, 163-184. CHOMSKY Noam & HALLE Morris, 1968, The sound pattern of English, New York: Harper and Row. COOPER William E. & ROSS John R., 1975, “World order”, in GROSSMAN Robin E., SAN James L. & VANCE Timothy J. (Eds.), Papers from the parasession on functionalism in linguistics, Chicago: Chicago Linguistics Society, 63-111. FRADIN Bernard, 2002, “Combining forms, blends and related phenomena.”, Extragrammatical and marginal morphology 12, 11‑59. GRIES Stefan Th., 2004, “Shouldn’t it be breakfunch? A quantitative analysis of blend structure in English”, Linguistics 42, 639-667. Lexis, 14 | 2019 91 GRIES Stefan Th., 2012, “Quantitative corpus data on blend formation: Psycho- and cognitive- linguistic perspectives”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre (Eds.), Cross- disciplinary perspectives on lexical blending, Berlin: De Gruyter Mouton, 145‑167. HASPELMATH Martin, 1995, “The growth of affixes in morphological reanalysis”, Yearbook of Morphology 1994, Dordrecht: Springer, 1-29. KIM Yuni, 2010, “Phonological and morphological conditions on affix order in Huave”, Morphology 20.1, 133-163. LEHRER Adrienne, 1996, “Identifying and interpreting blends: An experimental approach”, Cognitive Linguistics 7.4, 359-390. LEHRER Adrienne, 2007, “Blendalicious”, Lexical creativity, texts and contexts 19, 115‑136. LEVENSHTEIN Vladimir I., 1966, “Binary codes capable of correcting deletions, insertions, and reversals”, Soviet physics doklady 10(8), 707‑710. LÓPEZ RÚA Paula, 2004, “The categorical continuum of English blends”, English studies 85, 63-76. MILLER Gary D., 2014, English lexicogenesis, Oxford: Oxford University Press. MORETON Elliott, SMITH Jennifer L., PERTSOVA Katya, BROAD Rachel & PRICKETT Brandon, 2017, “Emergent positional privilege in novel English blends”, Language 93.2, 347‑380. PIÑEROS Carlos-Eduardo, 2004, “The creation of portmanteaus in the extragrammatical morphology of Spanish”, Probus 16.2, 203-240. PRINCE Alan & SMOLENSKY Paul, 2004 [1993], Optimality Theory: Constraint interaction in generative grammar, Oxford: Wiley-Blackwell. RENNER Vincent, 2008, “On the semantics of English coordinate compounds”, English Studies 89:5, 606-613. SCHOENFELD Aviv, forthcoming, “Secreted affixes in Biblical Hebrew personal names”, in DORON Edit & NOTARIUS Tania (Eds.), Brill’s Journal of Afroasiatic Languages and Linguistics. TOMASZEWICZ Ewa, 2008, “Novel words with final combining forms in English. A case for blends in word formation”, Poznań studies in contemporary linguistics 44.3, 363-378. WARREN Beatrice, 1990, “The importance of combining forms”, in DRESSLER Wolfgang U., LUSCHÜTZKY Hans C., PFEIFFER Oskar E. & RENNISON John R. (Eds.), Contemporary morphology, Berlin: Mouton de Gruyter, 111-132. WILLIAMS, Edwin, 1981, “On the notions “lexically related” and “head of a word””, Linguistic Inquiry 12, 245-274. APPENDIXES APPENDIX: Proportional string edit distance (PSED) The proportional string edit distance [Levenshtein 1966] runs between 0‑1. 0 is identity, and 1 is complete dissimilarity. Segment‑zero mismatches cost 1, linearity mismatches cost 1, and consonantal mismatches cost between 0.2–0.8, depending on number of differing distinctive features (place, manner, voicing and sonorant/obstruent), as detailed in Tables 10‑11, adapted from Bailey and Hahn [2005]. Similarity, vocalic Lexis, 14 | 2019 92 mismatches cost between 0.17–0.83, depending on number of differing features (high, low, back, round and ATR), as detailed in Tables 12‑13. Finally, the proportional string edit distance between two strings is the cost divided by the number of segments in the longer string. Table 10 : Consonantal features Table 11: Consonantal featural similarity Table 12: Vocalic features Lexis, 14 | 2019 93 Table 13: Vocalic featural similarity ABSTRACTS In this paper, we explore the conditions that result in variable base‑word positioning in English blends, where the same base words have variable order, yielding two blends, i.e. blend doublets. In non‑synonymous doublets, such as (egg × prégnant >) éggnant ‘pregnant with egg’ and prégegg ‘egg that counts down pregnancy’, the main factor is that in endocentric blends, the base word that contributes the semantic head is right‑aligned. In synonymous blend doublets, such as plúmpricot ~ ápriplum ‘plum‑apricot hybrid’ (< plum × ápricot), variable base‑word positioning results from at least four factors interacting: (i) segmental faithfulness – maximizing segmental similarity; (ii) Pāṇini’s law – positioning the short base word before the long one; (iii) monosyllabic integrity – keeping the monosyllabic base word in the same syllable; and (iv) syntagmatic faithfulness – matching the linear order of the base words to their order in a would‑be syntactic constituent. Cet article explore les conditions qui résultent du positionnement variable du terme source dans les amalgames anglais, dans lesquels les mêmes termes sources ont un ordre variable, créant ainsi deux amalgames différents, c’est-à-dire des doublets d’amalgames. Pour les doublets d’amalgames non synonymiques, tels que (egg × prégnant >) éggnant ‘pregnant with egg’ et prégegg ‘egg that counts down pregnancy’, le facteur principal est que dans les amalgames endocentriques, le terme source qui constitue la tête sémantique se place à droite. Pour les doublets d’amalgames synonymiques, tels que plúmpricot ~ ápriplum ‘plum‑apricot hybrid’ (< plum × ápricot), le positionnement variable du terme source dépend d’au moins quatre facteurs qui se combinent : (i) la fidélité segmentale – qui permet l’optimalisation de la similarité segmentale ; (ii) la loi de Pāṇini – qui consiste à placer le terme source le plus court devant le terme source le plus long ; (iii) l’intégrité monosyllabique – c’est-à-dire la conservation du terme source monosyllabique dans la même syllable ; et (iv) la fidélité syntagmatique – qui ajuste l’ordre linéaire des termes sources selon l’ordre qu’ils auraient dans un constituant syntaxique potentiel. INDEX Keywords: blends, variable base‑word positioning, variation, endocentric, exocentric, coordinative Mots-clés: amalgames, positionnement variable du terme source, variation, endocentrique, exocentrique, coordinatif Lexis, 14 | 2019 94 AUTHORS AVIV SCHOENFELD Tel Aviv University [email protected] EVAN GARY COHEN Tel Aviv University [email protected] OUTI BAT‑EL Tel Aviv University [email protected] Lexis, 14 | 2019 95 A study on the ‘wordgasm’: the nature of blends’ splinters Alejandro Barrena Jurado I would like to thank Carmen Portero Muñoz, senior Lecturer at the University of Cordoba, for instilling in me a great love of morphology through her lectures. Her supervision and advice, but most of all, her passion and dedication, were fundamental for me to write this paper. Introduction 1 The use of blending as a morphological word-formation process does not cease to increase through the years. Some reasons behind this reality are probably blends’ creative effects, which may be used to capture people’s attention, or the degree of recognizability of the bases, which makes it very easy for language users to recover them. What is clear is the fact that, today, it is rare to read an article in a newspaper or scroll down your Twitter feed without coming across at least one of these creations. 2 From a linguistic perspective, however, the process of blending is way more interesting than the acknowledgment of the speakers’ wittiness when they coin these words. In fact, the creation of blended words gives rise to what morphologists call ‘splinters’, that is, elements that are used recurrently in the formation of new words, such as ‘-oholic’ in shopoholic. The nature of splinters remains quite dubious. On the one hand, new blends containing them could be seen as coinages that take the same source word as an element of the lexeme. Following this approach, new words ending in ‘-oholic’ would be seen as blends which take alcoholic as their second source word. On the other hand, they might be treated as words that are formed by attaching a suffix to a base. Following this interpretation, we would be considering ‘splinters’ as affixes, that is, morphemic elements that acquire some meaning. In this way, new words containing ‘- oholic’ would not necessarily take the word ‘alcoholic’ itself as their second source word. Although this discussion remains open and it is difficult to support strongly one of these views, new approaches to grammar and morphology make it easier to account for the formation of these blends. One of these theories is Construction Morphology, Lexis, 14 | 2019 96 which holds the belief that words, just like any other elements in the language, are constructions, that is, pairings of form and meaning. 3 In any case, not every splinter is used with the same frequency. Some elements might be used in a limited number of blends, whereas some others could be used in hundreds of new words. Such is the case of ‘-gasm’, a sequence coming from the word orgasm, which seems to be the basis of a very large number of words. A deep study of the new words, as well as their semantics, might shed some light on the nature of this splinter. Furthermore, the use of constructional schemas appears to be quite appropriate when it comes to specifying the differences in use of this splinter. 1. Aims of the study 4 The aim of this paper is to study the nature of blends’ splinters, elements that have been often referred to as ‘final combining forms’, whose status remains unclear. This work presents a case study in order to corroborate the fact that they are productive units, which are used recurrently to create new words, even though the new lexemes including them might have a very short lifespan. Among the different theories that have been developed, the possibility of the new words having been created by analogy and the idea that they are coined by suffixation stand out. Assuming the second option was the real situation, splinters would actually need to acquire a morphemic status and a suffix-like behaviour. Nevertheless, some other theories, such as construction morphology, favour the role of schemas as adequate tools for accounting for the phenomenon described above. 5 The specific objective of this work is to study words containing the element ‘-gasm’ to check tendencies in their formation. Although productivity is quite difficult to calculate, this study attempts to prove that the sequence ‘-gasm’ possesses a high degree of productivity and is used recurrently in the formation of new words. Through the analysis of the bases to which the splinter is attached and the study of the relationship existing between the component parts of the blends, I intend to discover which mechanism is at work, while attempting to accommodate the operation within the theoretical framework of Construction Morphology. 2. Theoretical framework 2.1. Blending as a conceptual mechanism 6 The theory of conceptual blending, also known as conceptual integration, was originally developed by Gilles Fauconnier and Mark Turner [1998] on the basis of some fundamental concepts in the realm of cognitive linguistics, namely frames, domains and mental spaces. Before delving into the study of conceptual blending, it is therefore necessary to define these preliminary concepts which will be used throughout this work. 7 As Geerarts & Cuyckens [2007] point out, the notion of frame has not solely been used in linguistics, but also in some other fields, such as psychology and artificial intelligence. However, it has played a major role in cognitive linguistics, with Charles J. Fillmore being one of its most influential figures and exponents. Radden & Dirven Lexis, 14 | 2019 97 [2007: 9-10] claim that “[the] coherent ‘package of knowledge’ that surrounds a category and is activated when we use or hear a word is known as a conceptual frame […] all our coherent bits of knowledge are structured in conceptual frames”. Thus, frames could be characterized as the pieces of information evoked by words. They are essential to foster communication and involve our perception of the world. 8 Regarding the notion of conceptual domain, Radden & Dirven [2007: 11] state: [it] is the general field to which a category or frame belongs in a given situation. For example, a knife belongs to the domain of ‘eating’ when used for cutting bread on the breakfast table, but to the domain of ‘fighting’ when used as a weapon. Whereas frames are specific knowledge structures surrounding categories, conceptual domains are very general areas of conceptualization. Some typical domains are those of ‘space’, ‘time’, ‘emotion’, ‘sports’, ‘travelling’, etc. Conceptual domains crosscut with frames and thus allow us to link frames to one another. 9 Mental spaces are defined by Fauconnier & Turner [2002: 102] as follows: Small conceptual packets constructed as we think and talk, for purposes of local understanding and action. They are very partial assemblies containing elements, structured by frames and cognitive models… [They] operate in working memory but are built up partly by activating structures available from long-term memory. 10 They are described by Radden & Dirven [2007: 30-31] as “short-lived packages of knowledge evoked on-line in communication”. The scholars use this explanation as a starting point to introduce conceptual blending, which is described as “the integration of two or more spaces into a “blended space” [which] inherits partial structure from its input spaces and has emergent meaning of its own”. The consideration that the blended space has its own emergent meaning is supported by Ungerer & Schmid [1996: 259], who claim that the new blended space’s set up “differs from those of the two input spaces”. 11 Taylor [2012] follows the same definition and illustrates conceptual blending with the example of someone imagining how a rug in a shop would fit their living room in their house. By doing so, the customer is creating a blended space (the image of the rug in their living room) through the fusion of elements from two different input spaces (the rug in the shop, its original location, as the first mental space; and the living room, without the rug, as the second space). Another very important aspect of conceptual blending highlighted by Taylor is the fact that it is not a compositional operation, but rather a selectional one. Taylor [2012: 265] points out that “the process is not compositional, with one (mental) space being added to, or superimposed on, the other. Integration involves selective activation of elements from the different spaces, whereby discrepancies are overlooked and differences in time and space are compressed”. 12 In addition, the possibilities of conceptual blending are numerous, as the creation of a blended space is not only restricted to two input spaces. Fauconnier & Turner [2002: 8] argue that the process “can operate over any number of mental spaces as inputs. Blending can also apply repeatedly: The product of blending can become the input to a new operation of blending”. The previous statement underlines the recursive nature of the phenomenon subject to study. Lexis, 14 | 2019 98 2.2. Lexical blending 13 Within the study of conceptual blending, many different elements and processes can be found. One of them is lexical blending, frequently referred to as ‘morphological blending’. In fact, Ungerer & Schmid [1996: 268] argue that morphological blends are the best exponent to study conceptual blending. 14 The importance in use of these words has been a subject of debate throughout the years, with some authors claiming that lexical blending was not a common strategy in language at all, and therefore not a major morphological process. Nevertheless, some researchers disagree, including Cannon [2000], among others. Crystal [1995: 130] claims that “blending seems to have increased in popularity in the 1980s, being increasingly used in commercial and advertising contexts”, although he casts some doubt on the success of the words coined by this process, explaining that most of them might have a very short lifespan prior to their disappearance. Taylor [2012: 266] notes that these lexemes are “often employed to name new products, new institutions, and new practices and concepts”. What is clear is that the use of this process is becoming ever more common and anybody is likely to find new blends when reading online news or participating in social media. 15 Lexical blends are defined by Arndt-Lappe & Plag [2013: 541] as “two-constituent compounds in which at least one constituent has lost some phonological material, and in which the left, or initial, part of Word 1 and the right, or final, part of Word 2 survives”. The structural pattern described by these scholars should not, however, be taken as a defining rule of morphological blends, as many instances of words formed by this process can be found which do not take the first and second elements from the first and second source words, respectively. One example is Whatsnap, a term coined by fusing the first elements of both WhatsApp and Snapchat. The previous case could be seen as an instance of clipped compounding, since both bases are equally clipped and semantically equivalent, i.e. one is not subordinated to the other. However, I will restrict clipped compounding to cases in which an already existing compound word is shortened. 16 Beliaeva [2014: 30] proposes a similar definition of lexical blends, which encapsulates what previous scholars state about the process, but is not as problematic as regards their structure. She defines a blend as: a lexical item formed by merging together two (or more) source forms, so that 1) only part of their orthographical and/ or phonological material is preserved, and 2) they have not been formed by concatenation of morphs. 17 In terms of conceptual blending, Kemmer [2006: 71] provides a very insightful definition, characterizing blends as “words that are cognitively linked to pre-existing words which are co-activated when the blend is used”. This definition is highly concerned with the cognitive nature underlying this word-formation mechanism, and it is quite appropriate in order to understand lexical blending as a sub-type of a major, overarching category, that of conceptual blending. An important characteristic of blends is their iconic nature. As Fandrych [2008: 111] demonstrates, “the name ‘blending’ is metaphorical, as blends ‘mix’ random parts of existing lexemes (‘splinters’) – structurally and semantically – and there is the additional semantic component BLENDING/MIXTURE… their forms reflect their referents”. Lexis, 14 | 2019 99 18 Although many aspects of blends are still to be studied, they might be one of the most important grammatical resources of the language, since, according to Taylor [2012: 266] they “can even give rise to what appear to be new morphological resources”. One example is the emergence of new affixes that might become productive, as will be discussed in the subsequent section of this work. 19 The classification of blends has been diverse through the literature, and different taxonomies have been proposed. Pound [1914], for instance, distinguishes different categories depending on the origin or the cause of coinage of the words, whereas some other scholars categorize lexical blends in relation to their structure (see Algeo [1977]; Bauer [1983]; Beliaeva [2014]). 20 Some research studies (see Algeo [1997]; Gries [2004b]; Kemmer [2006]) take into consideration the presence or absence of phonemic or graphemic overlap, that is, the presence in the blended words of phonemic elements or graphemes that are shared by both the two source words. Algeo [1997: 56-57] makes another distinction, differentiating syntagmatic blends, “the combination of two forms that occur sequentially in the speech chain”, such as Chicagorilla (Chicago + gorilla) from associative blends, where the source words are “linked in the word-maker’s mind and thence in his language”, as in words like shill (shiver + shill). A similar distinction is made by Bauer [2006: 502-503], who classifies blends into syntagmatic or paradigmatic according to their origin. Paradigmatic origin blends are those where the source words “are in a paradigmatic relationship with each other”, as in smog ( smoke + fog), while in syntagmatic origin blends their source words are characterized by being “in a syntagmatic relationship to each other”, as it happens in motel (motor + hotel). Beliaeva [2014] classifies blends in purely structural terms, depending on the parts of each source word (initial, final) are kept in the blended word. Choroleeva [2015: 902] also distinguishes endocentric from exocentric blends, following a terminology that has long since been applied to classify compound words in English. According to her, in endocentric blends “the first component modifies the second one, the latter functioning as a semantic head. In this case, the two elements of the derived lexeme are in attributive relations”, whereas in exocentric blends “the derived form consists of components which are semantically on a par because both function as semantic heads”. Choroleeva’s distinction is thus equivalent to Bauer’s [2006]. 21 Similarity and recognizability are two fundamental characteristics in the formation of English lexical blends. The operation of conceptual integration can be easily identified in these types of coinages, since they are combinations of parts of the two source words, which might be clipped or overlapped but are not radically transformed. The source words are thus easily recoverable. In relation to this issue, Lehrer [1996: 366] claims that the degree of similarity between the blend and the source words has an impact in the levels of recognizability of the blend and in its understanding. He states that “the more material from the target word that is present, the easier the blend is to identify […] if one part of a blend is identified, its semantics will be relevant to identifying the other part”. In addition, Gries [2004b] explains that when coining blends, both source words are fused in such a way so as to remain recognizable and to be quite similar to the blend itself, as regards not only letters or graphemes, but also phonemes, length and prosodic aspects like stress pattern. 22 One of the internal aspects of blends which has received most attention has been the location of their switch-point, that is, the place in the blend where the transition from Lexis, 14 | 2019 100 the first to the second source words is found. However, no definite conclusions have been found and the generalizations proposed cannot be used to account for all the existing blends. The ideal situation in the case of blending is the existence of phonemic or graphemic overlap, so that the transition is quite natural. In these cases, identifying a specific location as the blend’s switch-point is an arduous task, as the material to be considered for that aim is shared by the two input words. Kelly [1998] notices that whole syllables tend to be preserved and the switch-point normally occurs within syllabic boundaries. Nevertheless, this is not always the case. Many scholars have used corpora to reach conclusions, but they have not reached an agreement. Even though their frequency varies, switch-points can be virtually found in any place within the syllabic constituents of the words constituting the blend. 23 Another interesting feature of blends which has been extensively studied is their stress. There are two main perspectives on this issue. The first one, proposed by Cannon [1986: 746] establishes a relationship between the stress of the blend and the length of the source words, with the longer word dictating the primary stress of the blend. The second approach to blend stress highlights the importance of the second source word, arguing that it is always its stress pattern which is kept in the blend (see Gries [2004a]; Arndt-Lappe and Plag [2013]). 2.3. Splinters 2.3.1. The evolution of the term 24 This section presents an overview of the notion of ‘splinter’, as well as a revision of the evolution of the concept through the literature. The term has been traditionally associated with the morphological process of lexical blending, as a label to identify what Bauer, Lieber & Plag [2013: 19] refer to as “a portion of a word that is non- morphemic to begin with but has been split off and used recurrently on new bases”. 25 The term ‘splinter’ was originally used by Berman in 1961 to refer to these elements of a dubious nature. Berman [1961: 279] introduced the label in his definition of blending as “a process of coining new words under which a blend is formed by adding the splinter of the last initial word to the stem or to the shortened substitute of the stem of the first initial word (words)”. It is important to note that Berman uses the phrase ‘initial word’ to refer to what has been named ‘source word’ in this work. It is interesting to consider the fact that his definition does not view blends as fusions of elements from two words, but rather as the addition of a portion of a word to the stem of another lexeme. In his description of the process, Berman somehow equates the nature of ‘splinters’ to that of suffixes. Nevertheless, the scholar’s depiction of ‘splinters’ remains quite vague. 26 Adam’s [1993: 142] treatment of the term is a little more informative. He makes it clear that these elements are neither morphemes nor ‘compound-elements’, stating that “usually splinters are irregular in form, that is, they are parts of morphs, though in some cases there is no formal irregularity, but a special relationship of meaning between the splinter and some ‘regular’ word in which it occurs”. 27 Soudanek [1998] expands on the notion of splinter, distinguishing two different types: initial and final splinters. Fandrych [2008: 112] explains this difference on the grounds of the location of the element in the blend, by claiming that: Lexis, 14 | 2019 101 Initial splinters may be the first or the second element, while final splinters can only become the second element of blends. Overlaps, for example, motel, often result from the merging of initial and final splinters. Splinters can even give rise to new morphological units through reanalysis. 28 Fandrych’s explanation includes some very interesting remarks that are relevant to any discussion of English blends. On the one hand, the scholar acknowledges the existence of initial splinters that could also be used recursively to create new blends, and the possibility that they can be the second element of the blended word, as in Whatsnap, a lexeme coined by fusing two initial splinters (coming from Whatsapp and Snapchat). On the other hand, the potential of these elements is emphasized, as Fandrych explains that the use of these elements could result in novel morphological units. 29 López-Rúa [2002: 37] carries out an exhaustive study on blends providing one of the most detailed definitions of splinters up to date. She defines them as: those graphic and phonemic sequences (not only in blends but also in peripheral initialisms) which are neither inflectional nor derivational morphemes, nor combining forms (electro-, -scope), and whose length generally allows their identification as belonging to a previous word. Consequently, splinters tend to be syllables or units larger than syllables in their sources, as Ox– and –bridge in Oxbridge (‘OXford and CamBRIDGE), or Digi– and –alt in Digiralt (‘DIGItal radar ALTimeter’). When they are shorter than syllables, their constituents are the syllable onset (i.e. the prevocalic consonant or consonants); the onset and the nucleus (prevocalic consonants + vowel); or the rhyme (vowel + postvocalic consonants or coda). 30 In her definition, she stresses the idea that these sequences do not possess a morphemic status. It is also interesting to note that she distinguishes splinters from combining forms, a term that has often been used interchangeably with the previous, as mentioned by Correia Saavedra [2014: 56]. 31 Bauer, Lieber & Plag’s [2013: 459] definition of splinters as “originally (mostly) non- morphemic portions of a word that have been split off and used in the formation of new words with a specific new meaning” does not add much to the literature, but summarizes briefly the main aspects of these sequences that have been previously discussed. 32 Beliaeva [2014: 32] clarifies some terminological issues by stating that: this term is used in at least two different senses. According to one interpretation, any “shorter substitutes” of words (Adams 1973: 142) should be called splinters. The other approach is to use this term only for those word parts that have started to be used productively in more than one blend, e.g. –(a)holic,–(a)nomics (Bauer 2006: 503). 33 Choroleeva’s [2015] work sheds new light on the topic, as she discusses the different approaches to the nature of splinters and the formation of English blends: one of them considers blends to be formed by suffixation, providing the elements subject to study with a morphemic status, whereas the other views blends as being created on the basis of analogical operations. In her work, Choroleeva [2015: 903] explains that: Some linguists believe that elements of the type are morpheme-like and may in the course of time become free morphemes like burger if their productivity increases. This means that the words containing such suffixoidal elements should not be treated as telescopic. According to other authors, however, blend words bring about blend words, not new morphemes. Lexis, 14 | 2019 102 2.3.2. Morphemization 34 As I stated before, one of the approaches to the nature of splinters is their consideration as suffixoids, that is, elements that function as morphemes and possess a morphemic status, such as ‘-gate’, attached to bases to form the words nipplegate or Monicagate. In that regard, splinters function in a similar way to suffixes, for they can be attached regularly to bases to coin new words. 35 Nevertheless, Lehrer [1998] makes it clear that not every blend creates a new morpheme, that is, not every final splinter in a blend becomes, or is likely to become, an institutionalized suffix. In fact, this situation is rare if compared to the huge number of blends that are created every day. According to Lehrer [1998: 4]: the creation of a blend does not necessarily result in a new combining form. It depends on whether the SPLINTER, that is, the truncated word, becomes productive, and since productivity is a matter of degree, there is a scale from highly productive morphemes like -holic to splinters that have been used only once (apparently). 36 Some years later, Lehrer [2007: 121] expands on the topic and explains that “when a splinter becomes so common that people start using it frequently, it may lose its connection with the source word and can be considered as morpheme in its own right”. 37 Lalic-Krstin [2014: 257-258] argues that lexical blending can lead to “morpho-semantic re-analysis and consequently (re)morphemization, which can then facilitate the production of whole series of what Bauer [1983: 96] calls ‘analogical formations’”. In her study, Lalic-Krstin [2014: 258] explains that “through recurrent blending a splinter can gain morphemic status. This happens as a result of morpho-semantic re-analysis during which a lexeme is reinterpreted and a phonological string that previously had no morpheme status is perceived as a constituent”. 2.3.3. Requisites for splinters to become morphemes 38 The idea that splinters may acquire morphemic status is only supported by some research studies, in favour of the approach that considers splinters to be suffixoids, as opposed to viewing blends as the output of analogical processes. For this reason, the precise moment when a splinter can be treated as a morpheme remains unclear. The most obvious indicator of their status is the splinter’s productivity. Nonetheless, as Lehrer [1998: 5] affirms, “whether and when a splinter becomes productive appears to depend on nonlinguistic, mostly chance factors”, which complicates things when it comes to analyzing these forms. 39 In fact, Lalic-Lrstin [2014: 258] follows some previous scholars and concludes that: there is some disagreement as to what are the prerequisites for a splinter like this to become a legitimate morpheme. Warren 1990 believes a new morpheme emerges as soon as it participates in the formation of a new blend, whereas Lehrer 1998 finds it necessary for the splinter to gain some frequency in blend formation in order to become a fully fledged morpheme. 40 Some scholars have attempted to quantify the necessary number of lexemes including a splinter to consider it a suffixoid. Among them, Fischer [1998: 65] claims that there might be at least three neologisms. Nevertheless, a consensus has not been reached, Lexis, 14 | 2019 103 either for the number of neologisms or for the fact that frequency is an adequate measure tool to provide splinters with such a status. 41 One of the latest efforts to account for the morphemization of splinters has been made by Correia Saavedra [2014], who has developed an automated method based on a script capable of processing huge amounts of data and to signal which splinters are good morpheme candidates. Nevertheless, the method stills needs to be revisited and updated to allow for an unsupervized analysis of splinters and their classification as morphemes. 2.4. Blends within a theory 42 As Kemmer [2006: 7-8] observes, “despite the interest they attract due to their creative nature and their typological unusualness, lexical blends have been little more than a footnote in the study of morphology in modern linguistics”. She interestingly notes that “the amount of similar structure (in blends) can vary a great deal, so it is impossible to state a general formal rule that will license some blends and exclude others”. This peculiarity of blends, whose nature can only be described by positing tendencies, does not find a place in traditional rule-based approaches to grammar. In fact, Kemmer [2006: 9] argues that “it seems as though, instead of rules, speakers are operating with a facility for global pattern-matching that allows similarities on many different dimensions to count, as long as there are enough of them (with “enough” not being precisely quantifiable)”. This vision may be linked to the fields of paradigmatic morphology and construction grammar. Taylor [2012: 263] favours this interpretation of blends, by stating that: blending turns out to be a potent source of new expressions and new constructions and thus offers itself as a serious alternative to the rule-based mechanisms of generative theory […] Previously encountered expressions, and the schematic constructions that they instantiate, can be the source of new linguistic expressions. Speakers can introduce variations to performed expressions, they can extend accepted usage patterns, and they can blend existing resources to create hybrid expressions which inherit some aspect of their inputs. 2.4.1. Paradigmatic morphology 43 Most morphological processes have traditionally been described from a syntagmatic perspective, based on the notion of concatenation (or the attachment of elements to bases) or from a paradigmatic point of view, considering groups of lexemes to be related to one another [Bauer, Lieber & Plag 2013]. 44 Bauer, Lieber & Plag [2013] argue that every word is part of a paradigmatic morphological relation with other words sharing the same base or suffix. They use the label “morphological category” to categorize words containing the same affix and “morphological family” to include words which share the same base. Many experiments, mainly lexical decision tasks, have been carried out by researchers to prove the psycholinguistic reality of the previous concepts, as detailed by Schreuder & Baayen [1997] and Plag & Baayen [2009]. Lexis, 14 | 2019 104 45 In any case, paradigmatic morphology is not restricted to the previous cases. It is also concerned with the study of words that are orthographically, phonetically or semantically related. 46 The concept of analogy has often been used by scholars, like Becker [1993] or Ardnt- Lappe [2014] to illustrate cases where a new lexical unit is coined through comparison with other words, as in the creation of deverbal adjectives ending in ‘-able’. Other authors, like Burzio [1998], Steriade [2000] or Raffelsiefen [2004] use the terms ‘multiple correspondence’, ‘paradigmatic uniformity’ or ‘stem selection’, respectively, to make reference to complex words with phonological properties influenced by groups of related words. Bauer [2001: 76] describes that the principle of analogy in derivation dictates that any new derivative is created provided there is “a suitable pattern for it to be formed on”. 47 The process of blending, whose main characteristics have been described above, is hard to describe from a syntagmatic point of view, and thus paradigmatic approaches seem to be able to provide a more plausible explanation of it. Kemmer [2006] is one of the first scholars to work on blending using this approach and explaining that rule-based morphology is not adequate for a correct study of these formations. In Kemmer [2006: 9], the scholar constructs her theory on the basis of the concept of schema, which she defines as “a cognitive representation consisting of perceived similarities across many instances of usage. Schemas are essentially routinized, or cognitively entrenched, patterns of experience”. The researcher provides a detailed explanation of schemas, signalling their most interesting features and their cognitive basis, and linking their emergence to productivity. Kemmer [2006: 10] indicates that: Schemas, being simply what two or more forms or meanings have in common, are not restricted by the kinds of sequencing required by rules, nor are they restricted to information of particular types or levels. Schemas allow for competing factors in licensing expressions, and for multiple possible sanctioning structures, thus giving rise to variation and to possible non-unique analyses for linguistic expressions. And, importantly, schemas are sensitive to frequency: frequency measures repetition of instances, which leads to conventionalization (cognitive entrenchment and degree of dispersal in a group of speakers) which itself affects the possibilities licensed by the system: instances of use, over time, shape the very system that gives rise to them. 48 In addition, recent grammatical models belonging to Construction Grammar have paved the way for the study of blends as constructions. 2.4.2. Construction Grammar 49 Constructionist approaches take the notion of construction as their basic elements. As explained by Goldberg [2006], they understand constructions as autonomous abstract entities which are perceived to be psychologically real by language users. These approaches also hold the idea that constructions exist independently of the lexical items that fill them. According to these approaches, any type of lexicogrammatical unit, such as a morpheme, a word, or an idiom, is considered to be a construction, that is, a pairing of form and meaning. 50 Construction grammar emerges as a reaction against Chomsky’s generative grammar, and deviates from it by positing a number of considerations. One of the main views of this trend is that the totality of our knowledge about language is structured in Lexis, 14 | 2019 105 interrelated constructions. Construction grammarians oppose to Chomsky’s belief that grammar and lexicon are two separate modules. Goldberg [1995: 7] argues that “lexical constructions and syntactic constructions differ in internal complexity […] but both lexical and syntactic constructions […] pair form and meaning”. Construction grammars also consider semantics and pragmatics to be part of a continuum, without clear-cut boundaries. 51 Nevertheless, Goldberg [1995: 7] highlights the generative nature of construction grammar “in the sense that it tries to account for the infinite number of expressions that are allowed by the grammar while attempting to account for the fact that an infinite number of expressions are ruled out or disallowed”. Goldberg [1998: 205] defines constructions as follows: “C is a CONSTRUCTION iff def n C is a form-function pair, such that some aspect of the form or some aspect of the function is not strictly predictable from C’s component parts”. Later on, in Goldberg [2002: 813], she defines a construction as “a pairing of form with meaning/use such that some aspect of the meaning/use is not strictly predictable from the component parts or from other constructions already established as existing in the language”. 2.4.3. Construction Morphology 52 Construction Morphology is an approach to morphology within the overarching theory of Construction Grammar. It was developed by Dutch linguist Geert Booij. Construction Grammar is not solely a theory of syntax and the notion of construction is also relevant to the study of morphology. Words, being essentially pairings of form and meaning, are an instance of constructions. In fact, Goldberg [2006: 18] specifies that “the network of constructions captures our grammatical knowledge of languages in toto, i.e. it’s constructions all the way down”. 53 Construction morphology is a usage-based approach. As explained by Booij [2015: 425], “language users first acquire words, and only once they have acquired a sufficiently large set of words of a certain type can they conclude to abstract morphological patterns”. This is an important idea, as it implies that this approach to morphology rejects the rule-list fallacy, or what Booij [2010: 4] identifies as “the unwarranted assumption that linguistic constructs are either generated by rule or listed, and that being listed excludes a linguistic construct from being linked to a rule at the same time”. Booij actually defends the idea that abstractions allowing speakers to coin new complex words are based on actual instances of words that are memorized. 54 Within morphological constructions, we can find both inflectional and derivational cases. Some examples of inflectional constructions are the plural construction (which has unpredictable form and meaning, as well as idiosyncratic properties) or the past tense construction. An instance of a derivational construction is that which enables us to create adjectives ending in ‘-able’, as in ‘drinkable’. We can say that word formation processes correspond to linguistic generalizations. The generalization allowing us to interpret and create new cases is a construction. Regarding lexical blending, there are formal generalizations in the minds of speakers. 55 Morphological constructions are represented by means of constructional schemas in this approach to grammar. Some examples, taken from Booij [2015], are the following: (1) <[x]Vi er]Nj ↔ [Agent of SEMi]j> (2) <[x]Ni less]Aj ↔ [Property of being without SEMi]j> Lexis, 14 | 2019 106 56 In his work, Booij [2015: 425] explains that: In these schemas, the double arrow indicates the correlation between form and meaning. By means of co-indexation the systematic relationship between form and meaning is specified. The index i in these examples serves to indicate that the meaning of the base word (SEM) recurs in that of the corresponding complex word. The index j indicates that the meaning of the construction as a whole correlates with the form as a whole. The angled brackets demarcate a constructional schema. 57 It is important to distinguish constructions from constructs. While the former refers to abstract schemas, the latter makes reference to instantiations, which are, according to Booij [2015: 431] “individual words that inherit the information specified in the schema, thus making parts of the information contained in the lexical entries for these words redundant”. Abstract schemas and their instantiations are both regarded as constructions, but they exhibit different degrees of abstraction. 58 Booij [2015: 431] also introduces the notion of subschemas which make it possible to “express generalizations about subsets of complex words”. In the case of compounds, for instance, there are constructions with different degrees of complexity. The endocentric compound construction, for example, has several subschemas, such as the NN compound construction or the Top-N compound construction, a subschema underlying instantiations like ‘top-achievement’, ‘top-experience’. 59 An important issue as regards morphological constructions is productivity. Hilpert [2014: 81] argues that “the productivity of a schematic morphological construction describes the degree of cognitive ease with which speakers can produce or process new complex words on the basis of that construction”. Productivity is a gradient notion, and it is quite difficult to measure. With the aid of corpora, we can estimate the degree of productivity of a construction through type frequency and token frequency. Type frequency refers to the number of instantiations of a schematic morphological construction. In the case of blends containing the same splinter or affixoid element, type frequency would refer to the number of words containing that specific element. Token frequency, on the other hand, refers to the number of times that the same word is found within a corpus, that is, the number of different examples found for a single word. In the study of token frequency, hapax legomena stand as a clear indicator of high productivity, these being cases where there is only one token of a type, the lowest possible frequency [Plag 2003: 54]. Not all hapaxes are neologisms, but the existence of many may be a clue of the cognitive ease needed to create new words on the basis of a constructional schema. The more instantiations found, the higher the productivity of the schema. 3. Methodology 60 The first step in the development of this work has been the creation of a corpus containing words including the sequence ‘-gasm’. In order to do so, three different online corpora have been used: the Global Web-Based English (GloWbE, available at https://corpus.byu.edu/glowbe/), the Corpus of Contemporary American English (COCA, available at https://corpus.byu.edu/coca/) and the English Web Corpus (EnTenTen2015, accessible through the corpus management system SketchEngine: https:// www.sketchengine.eu/). The main reason for choosing these corpora has been their size: EnTenTen2015 contains over 18 billion words, GloWbE includes more than 1.9 Lexis, 14 | 2019 107 billion, and COCA is composed of 560 million words, making them three of the largest online collections of words. The longevity of each of the corpora has been another important factor, for the three of them have been updated and new words have been added to them on a regular basis. In addition, another determining factor for choosing the three previous tools has been their heterogeneous nature, as they comprise words coming from a wide variety of sources. The GloWbE corpus, for instance, includes texts from twenty different countries. Furthermore, all of them are made up of different types of texts, from transcribed spoken pieces of language to academic texts. Nevertheless, some other opportunistic sources have been used. The decision to do so was based on the nature of the process of blending itself: due to the creative effects that can be reached by using it, blends are generated with a very high frequency. The majority of blends, however, do not become very popular and they are not often included in the texts that form part of online language corpora, but rather they are commonly found in social media. For that reason, some blends ending in ‘-gasm’ have been taken from Twitter, blogs and newspapers’ websites and even online discussion threads consulted randomly. Some examples of blends coming from these sources are Gagasm ( Gaga + (or)gasm, meaning ‘a state of excitement or pleasure produced by listening to music by Lady Gaga) or Swirllgasm (Swirll + (or)gasm, meaning ‘a state of extreme pleasure as a result of eating Swirll yoghurt). 61 The same procedure was followed to extract words containing the element ‘-gasm’ from GloWbE and COCA. I introduced the sequence “*gasm” under the List label, in order to obtain all the words where any type of material preceded ‘-gasm’. I changed the search options to allow the corpora to display up to 500 results, in case there were more words than expected containing the suffixoid element. Additionally, the search options were set to obtain all words with a minimum frequency of one token. Given the nature of the process of blending itself, it is often difficult to find many instances of a single word. In fact, the existence of hapax legomena (only one token for word in a corpus) is an indicator of the productivity of the process, which is a very useful point of departure for this work. As regards the EnTenTen2015 corpus, I accessed it through SketchEngine and searched for the words using the Word List tool. I changed the default filter options to only list words containing ‘gasm’ by introducing the command “.*gasm”, and I changed the minimum frequency of appearance to one token, for the reasons explained above. Once they had been retrieved, the words were revised in order to avoid any confusion, since the corpus tools occasionally provided variations of the same word, mainly linked to their spellings, with many lexemes written as the hyphenated variants of others. Such is the case of geek-gasm, which was listed as a different word from geekgasm. In these cases, the different variants were subsumed under a single lexeme, and the number of tokens of each were put together. Moreover, some words were discarded, since they lacked a context in the corpora or the context was a sequence of seemingly random words that showed signs of having been automatically coined by advertisement tools. Some other words containing neoclassical elements, like ‘mega’ or ‘micro’, were also excluded, as they remain closer to affixation in nature than to lexical blending, although others, such as cyborgasm, were kept, due to their dubious nature (it could have taken “cybernetic” as its source word, as opposed to the neoclassical element ‘cyber’ on its own). After following the aforementioned steps, 131 words ending in ‘-gasm’ were obtained from EnTenTen2015. GloWbE delivered 61 words, whereas COCA provided me with 13. On many occasions, the same word was listed in the three corpora, so the total amount of different words containing ‘-gasm’, taking Lexis, 14 | 2019 108 into account the combined results of the three corpora, is 173. Six more words were taken from the other sources, which added to the rest made up a corpus of 179 neologisms. In order to measure their productivity, the words extracted from the online corpora were listed in a table, to quantify the number of tokens for each and reach any conclusions. 62 Subsequently, the source words of all the blends subject to study were identified and a structural analysis was carried out with the intention of determining the type of bases to which ‘-gasm’ is attached, as well as the possible variations of the sequence ‘-gasm’. The number of syllables of the first source words and the resulting new blends were computed. As regards the semantic analysis of the output words, several factors were taken into account. The first stage in their classification was to contextualize the words. For that aim, an instance of each of the words used in a wider context was recovered using the corpus tools that enabled the compilation of the corpus. When no context could be recovered (in cases where the word appeared in isolation, or was a part of an advertisement, among other reasons), I used other resources, such as Urban Dictionary (https://www.urbandictionary.com), an online dictionary where users can upload new words, to obtain definitions or sentences including the words I was interested in. Additionally, a definition for each of the words was searched. The source of each definition is listed with the examples. When no definition could be retrieved from the web, I provided my own definition of the word ending in ‘-gasm’ based on the context. 63 Taking into account the source words of the blends and the context in which they appeared, I proceeded to analyze them in terms of their semantics, according to the relationships existing between the first and the second component of the novel coinages. The relationships between the source words of the blends were studied according to different factors, such as the notion of causality, the syntactic behaviour of the source words and the semantic roles fulfilled by the said source words. All the words that exhibited a similar behaviour were grouped under the same label and a constructional approach based on the existence of schemas and subschemas was used to account for these formations. The use of schemas to explain the creation of the blends was based on Booij’s [2010] work. 4. Analysis and results 4.1. Productivity of the process 64 In relation to the specific objective of this study, an approximation to the degree of productivity of the element ‘-gasm’ has been calculated. As we described in the revision of the literature, the productivity of a given process can be measured by taking into account type frequency and token frequency. Table 1 in Appendix 1 shows the relevant frequencies and the distribution of the words in the three corpora used: EnTenTen2015, GloWbE and COCA. 65 The combined type frequency of words ending in ‘-gasm’ is 173. As regards the token frequency of each type, it varies depending on the corpus used. However, one of the most significant aspects of Table 1 is the huge gap in token frequency between the word orgasm and the rest of the source words. We can conclude, therefore, that the rest of words take the lexeme orgasm as their point of departure. It is unclear from the data, Lexis, 14 | 2019 109 however, if all the words have been directly modelled upon orgasm or whether some of them have been coined by analogy with other words ending in ‘-gasm’ different from orgasm. The present case study suggests the idea that new lexemes are in fact created taking other blends ending in ‘-gasm’ as their input. The notion of schemas and subschemas might also be potentially used to account for these creations. For the aims of this study, the words whose token frequency is quite low are very interesting, especially those whose frequency is of one word only. These are hapax legomena. If we consider each of the three corpora separately, we find that there are 55 hapax legomena in the EnTenTen2015, 34 in GloWbE, and 11 in COCA. If we only take into account the words which are simultaneously instances of hapax legomena in the three corpora (or in one of them, if the new lexeme is not included in the others), we find 88 cases. As we discussed above, hapax legomena are an indicator of high productivity. The existence of such a big number of hapax legomena for words ending in ‘-gasm’ is therefore an indicator that this splinter is quite productive, as it is attached very regularly to new bases to coin new words. This finding supports several approaches to the nature of splinters. On the one hand, it fits the idea that these elements are subject to morphemization. The sequence ‘-gasm’ might be interpreted as an emergent derivational morpheme, as its occurrence with a wide variety of bases has been proven to be possible. On the other hand, the data also supports the appropriateness of using a constructional approach to account for the nature of these elements. The existence of many hapax legomena may indicate that a schema based on actual words containing ‘- gasm’ has been developed in the mind of speakers. 66 The previous results might serve as a possible indicator of the autonomy of the splinter ‘-gasm’, which could be used to create new blends without directly taking orgasm as their second source word. 4.2. Structural properties of the blended words 4.2.1. Variations of ‘-gasm’ 67 This section is concerned with the analysis of the structural properties of the words in my corpus, which have been created on the basis of orgasm. The first thing that has drawn my attention in their study has been the fact that there are some variations of the splinter used to coin new words. Although the main realization of this sequence is ‘- gasm’, this splinter changes to ‘-ogasm’ on some occasions. A few words are formed by attaching the whole form orgasm, instead of the back-clipped sequence ‘-gasm’, and others contain the sequence ‘-agasm’. 68 The words taking orgasm to create new words seem to have been directly coined on the basis of the original word, in its full form. The majority of them are cases where we find both graphemic and phonetic overlap, that is, some material is shared by both the first source word involved in the blend and the term orgasm. Such is the case of carnivorgasm (carnivore + orgasm), explorgasm (explorer + orgasm), floorgasm (floor + orgasm), gorgasm (gore + orgasm), horrorgasm (horror + orgasm), horgasm (horticulture + orgasm), morgasm (more + orgasm), prorgasm (programmer + orgasm), schizorgasm (schizophrenia + orgasm), and snoorgasm (snooze + orgasm). However, we also find some cases of blends including the whole word ‘orgasm’ where there is no overlap. These words are cyborgasm (possibly cybernetic + orgasm), femaleorgasm (female + orgasm) and Obamaorgasm (Obama + orgasm). The reason why these words take the whole form of the second source word is Lexis, 14 | 2019 110 unclear, especially in the case of the two last words, for which the alternative versions femgasm and Obamagasm can be found, respectively. One possible explanation is that the speaker who has coined these words wants to highlight the fact that the new blends are directly linked to the word orgasm. All these words, therefore, seem to have been created as novel blends, taking orgasm as their second source word, as opposed to being formed by the addition of a splinter. 69 The group of words containing ‘-agasm’ instead of ‘-gasm’ do not serve as an indicator of the existence of ‘-agasm’ as a variant of ‘-gasm’, for the grapheme ‘a’ is part of each of the source words. They simply reflect the situation in which ‘-gasm’ is attached to clipped versions of source words ending in ‘-a’ before the new blends are formed. Seven words conform this group: Berbagasm (Berbatov + -gasm), cameragasm (camera + -gasm), exploragasm (exploration + -gasm), extravagasm (extravagant + -gasm), ideagasm (idea + - gasm), Obamagasm (Obama + -gasm), and operagasm (opera + -gasm). 70 However, ‘-ogasm’ seems to behave like a real variant of ‘-gasm’ in some words, like luxogasm (luxury + -ogasm) or delish-o-gasm (delish + -o-gasm). Given the context in which ‘-ogasm’ is attached in these cases, one possible explanation for this variant is the need for vocalic support, as the splinter ‘-gasm’ is attached to a (sporadically) clipped base consisting of several final consonants. The addition of the vowel ‘o’ before the splinter could be due to the attempt to avoid the creation of a very long consonantic cluster. Another possibility is the use of this linking element by analogy with many neoclassical compounds. 4.2.2. Blend length 71 As regards the length of the words ending in ‘-gasm’, two things have been taken into account, namely the length of the bases to which the splinter is attached (the first source word of all the new blends) and the length of the novel blends themselves. 72 In relation to the length of the first source words, ‘-gasm’ is combined with bases of one syllable in 54.50% of the cases (97 out of 178 words), as in artgasm (the first source word being ‘art’). It is attached to 50 bases of two syllables, as in papergasm (accounting for a 28.09% of the cases), 23 bases of three syllables, as in Obamagasm (12.92% of the total), 7 bases of four syllables, as in literarygasm (3.93%) and one base of five syllables (0.56%), electrogasm, whose first source word is ‘electricity’. 73 Concerning the length of the new blends, we find 106 new lexemes with two syllables, like eargasm, accounting for a 59.55% of the corpus, 51 three-syllable blends, such as drawergasm (28.65%), and 19 words with four syllables, like cameragasm (10.67% of the cases). In addition to these preferred patterns, we also find 2 blends with five syllables, literarygasm and Obamaorgasm (accounting for 1.12% of the cases). 74 The previous distribution is not surprising at all, for, as I mentioned in the revision of the literature, one of the most important aspects of lexical blending is the degree of similarity between the blend and its source words, as well as the recognizability of the source words that constitute a blend. These results, consequently, go in accordance with previous research by Lehrer [1996] and Gries [2004b]. As mentioned above, the most frequent pattern for words ending in ‘–gasm’ is that where the sequence is added to a monosyllabic case to create a disyllabic word. This makes perfect sense, and it is, in fact, a rather ideal situation, since these output blends preserve their first source words in their totality, making them highly recognizable. Furthermore, by using monosyllabic Lexis, 14 | 2019 111 words as the first element in the blends, speakers mirror the length of the original word orgasm, and reproduce the syllabic pattern in their new creations. 4.3. Semantic properties of the blended words 75 The analysis of the relationship existing between the two elements in each blend (the first source word and the sequence ‘-gasm’, coming from orgasm) has enabled me to distinguish different categories whose characteristics I will present below. Although some of the words in the corpus do not fit into any of these groups, given their lack of a sufficient context so as to determine the relation between the words in a precise way, the majority of the blends can be classified using these categories, namely cause-and- effect blends, experiencer blends, coordination blends, predication blends and adjectival blends. All these groups are described in this section, and detailed examples of each group are provided. 4.3.1. Cause and effect blends 76 I have labelled this group of words as ‘cause and effect blends’ because the relationship between the two source words in each blend is based on this notion. All the lexemes contained in this group follow the same pattern: the first source word is used as the cause which provokes an effect in a person (an orgasm, a feeling of pleasure or excitement). On the basis of the study conducted in this work, most words ending in ‘- gasm’ seem to belong to this category, since cause and effect appears to be the most common relation between their constituents. 125 words out of 178 in the corpus exhibit this property, accounting for 70.22% of the total. The first source word of these blends is always a noun. We only find 4 cases which do not conform to this generalization, namely chillgasm (whose first source is a verb), cutegasm (adjective), gamingasm (verb) and scoregasm (verb). The nature of the nouns involved in the formation of the blends, however, is varied, including common nouns, such as food in foodgasm, proper nouns, such as Bieber in Bievergasm or abstract nouns, like fear in feargasm. 77 Additionally, another characteristic of this group of blends is the fact that their first source tends to be used metonymically, that is, standing for something else, as will be explained below. This property can be seen in words like heartgasm, where ‘heart’ does not stand for a body organ, but rather for ‘love’, or bookgasm, where ‘book’ does not stand for an actual copy of a book, but for the activity of reading a book instead. 78 An exhaustive semantic analysis of the words in this category has showed that two more specific groups can be distinguished, based on the specificity in meaning of the sequence ‘-gasm’. Although this constituent expresses an effect in all of the blends, we may find a slight variation in the meaning that the original second source word orgasm takes in the resulting blends. Thus, one group of words seem to replicate the original meaning of orgasm. In these words, the second element, which designates the effect, is used with the meaning ‘physical or bodily pleasure’. In a second group of blends the meaning of ‘-gasm’ appears to deviate from the original meaning In these words, the second element of the blend means ‘a feeling of excitement or enthusiasm’. Lexis, 14 | 2019 112 4.3.1.1. ‘-gasm’ meaning ‘physical or bodily pleasure’ 79 As has been mentioned above, the words in these groups follow the cause-and-effect pattern, where ‘-gasm’ is the effect and stands for the original meaning of orgasm, that is, physical or bodily pleasure (related to, or provoked by the first element of the blend). 80 The theory of Construction Morphology, proposed by Geert Booij, seems to provide the necessary tools to explain the formation of these words based on the idea of schemas. In fact, the creation of this group of blends may be explained by means of the following schema: <[[X]Ni gasm]Nj ↔ [A sensation of physical or bodily pleasure produced by SEMi]j> 81 It is important to note that the first word in this schema is also semantically variable, as in some cases it consists of concrete entity (footgasm) while in others, it is an activity (yogasm). 82 The majority of words in the previous group seem to be metonymic. More precisely, it is in the first source word where metonymy appears to be at work. The most common metonymy in these cases is that of a participant for the action or the activity in which that participant is involved. This can be seen in cases like footgasm, where foot has the semantic role of patient or locative, participating in the action ‘massage on foot’. Metonymy can also be considered to be at work in cases like yogasm, although it is not as evident as in footgasm, since the first source word of the former refers directly to an activity, that is, the degree of contiguity between the concepts is higher. 83 The word foodgasm, included in the previous list, illustrates another group of words that share a semantic property. Foodgasm makes reference to the feeling of pleasure one gets when eating something. An example of the word in use is the following: “One of the diners at an adjacent table had a noisy foodgasm over this dessert, so we tried it on our second visit” (www.magoguide.net). (EnTenTen15). 84 Using the theoretical framework of Construction Morphology, the subschema resulting from the word foodgasm can be described as follows: <[[x]Ni gasm]Nj ↔ [A sensation of physical or bodily pleasure produced by eating or drinking SEMi]j> 4.3.1.2. ‘-gasm’ meaning ‘feeling of excitement or enthusiasm’ 85 In a different group of words in the corpus, the sequence ‘-gasm’ does not reproduce the original meaning of orgasm, but a slight variation. This element is used with the meaning of ‘a feeling of excitement or enthusiasm’, produced by the first element of the blend. 86 Following the framework of Construction Morphology, the following schema can be used to account for these coinages: <[[X]Ni gasm]Nj ↔ [A feeling of excitement or enthusiasm produced by SEMi]j> 87 In addition, these words have something in common with the blends in the previous group: metonymy is also at work in the first element of many of these words. In many cases, therefore, the feeling of excitement is not caused by the entity denoted by the first word, but by the activity in which that entity participates instead. In TVgasm, for instance, the feeling of excitement is not caused by a television as an object, but rather by the activity of watching something one loves on television. Lexis, 14 | 2019 113 88 As in the group of blends where ‘-gasm’ means ‘physical or bodily pleasure’, in this second group, where ‘-gasm’ stands for a ‘feeling of excitement or enthusiasm’, we find a case which is semantically more specific and has probably paved the way for subsequent creations. 89 Eargasm makes reference to a feeling of maximal enthusiasm or excitement experienced when listening to music. The word audiogasm can also be considered as the origin of other creations, as it stands for the feeling of excitement experienced when listening to an audio track one particularly loves. 90 Using the principles of Construction Morphology, the new subschema can be accounted for as follows: <[[X]Ni gasm]Nj ↔ [A feeling of excitement or enthusiasm produced by listening to music by / the sound of SEMi]j> 91 This second order schema presents a higher degree of specificity at the semantic level. All the words created on the basis of this subschema maintain the meaning of the general schema, while expanding on it and specifying that the feeling of enthusiasm is produced by the activity of listening to a specific type of music, the sound of an instrument or a song by a singer. 4.3.2. Experiencer blends 92 The reason why I have labelled this group of blends as ‘experiencer blends’ is that the relationship between their source words is such that the first source word has the semantic role of experiencer, that is, an entity that is aware of or is affected by the action or state denoted by a predicate, the second source word. It is important to draw a distinction between an agent and an experiencer at this point, since the agent is the participant that instigates an action. For this reason, the first elements of these blends should not be confused with elements fulfilling the role of agent, for they are entities that are affected by the feeling of pleasure described by the second element, but do not have control of it. As expected, the first source words of all the words that constitute this group are nouns (some of them proper names). These blends can be paraphrased as “(an) X has/experiences an orgasm/pleasure”. 93 The cases of experiencer blends in our corpus account for 11.80% of the data, with 21 words characterized by this feature out of the 178 subject to study. Although they form a small group if compared with ‘cause and effect’ blends, they are part of the second biggest group that has been identified. 94 Following the theory of Construction Morphology, a general schema for the creation of words that follow this pattern could be represented as follows: <[[X]Ni gasm]Nj ↔ [A feeling of pleasure or excitement experienced by SEMi]j> 95 Lower-level schemas also seem to operate in the case of experiencer blends, as we find two clearly distinguished groups: one of them consisting of blends with initial proper nouns, and another one with common nouns. These subschemas, which differ in the semantics of the first element constituting the blend, are represented below. <[[X]PNi gasm]Nj ↔ [A feeling of pleasure or excitement experienced by SEMi]j> 96 In this case, PN stands for the semantic specificity of the nouns that are used to create the new words, that is, proper nouns. <[[X]CNi gasm]Nj ↔ [A feeling of pleasure or excitement experienced by SEMi]j> Lexis, 14 | 2019 114 97 In the second case, CN stands for common noun, as the first source word of the formations coined through this schema are semantically different from the ones which originate from the previous subschema. 4.3.3. Coordination blends 98 This group has been named ‘coordination blends’ because the source words of the blends stand in a relationship of coordination. This group is restricted to only 7 words out of the 178 in the corpus, accounting for a 3.93% of the total. Despite its small size, it is an interesting pattern, as it consists of words which describe the situation in which a feeling of pleasure or excitement and an additional sensation are experienced together by a person. 99 Following the principles of Construction Morphology, the schema explaining the creation of words of this nature is expressed below. <[[X]Ni gasm]Nj ↔ [A feeling of pleasure or excitement experienced together with (a feeling of) SEMi]j> 4.3.4. Adjectival blends 100 Although this group is not as clear-cut as the three previous categories, we find some words that have been formed by blending an adjective as their first source word and the word orgasm. In these cases, the word orgasm seems to maintain its full original meaning, or at least it is acknowledged as the direct second source word for the blend, and the first source words behave as premodifiers, attributing some qualities to the feeling of pleasure experienced by a person. In fact, these blends seem to behave like phrases and can be written as such. For that reason, all these words seem to have a semantic head, corresponding to the second source word. Six cases of words following this pattern can be found in the corpus, accounting for a 3.37% of the total. 101 A general schema explaining the formation of the blends could be the following: <[[X]Ai gasm]Nj ↔ [An orgasm of the nature described by SEMi]j> 4.3.5. Predication blends 102 Only one word has been found in the corpus to support the existence of this group. For that reason, it is unclear whether more words have been formed following this pattern or if some others will be coined in the future. However, the case found in the data is quite interesting, as the first source word is used as a predicate of orgasm. The word that exhibits this characteristic is artgasm. We could simplify the analysis in the absence of a surrounding context by saying that this word is an instance of a ‘cause and effect’ blend, where ‘art’ stands for ‘the activity of admiring a work of art’. However, the context where the word appears favours a different interpretation. (3) Artgasm. “The paper presents the Artgasm project that stimulates its participants into orgasm. The orgasm was physically induced to volunteer participants through a specialized, medical crafted “orgasmotron” and mediated to the audience through real time video, projected onto a large screen outside the performance space. With the help of qualified medical personnel, the project manipulates its male audience/participants to experience the maximum corporal pleasure, i.e. orgasm. In this manner, the Lexis, 14 | 2019 115 Artgasm project literally presents orgasms as both an ultimate form of art and aesthetic experience” (www.teatrutv.ubbcluj.ro). (EnTenTen15). 103 As can be observed, the meaning of the word when used in this context is ‘the consideration of an orgasm as being a work of art’. Furthermore, the blend could be paraphrased as ‘An orgasm is art’, where we can see very clearly that ‘art’ is a subject predicate, attributing qualities to orgasm. 104 Some other words in the corpus have been left out. The reason for this has been the difficulty at establishing the relationship between their source words. On many occasions, this difficulty is due to the lack of an appropriate context. 4.4. Final overview 105 This section aims at summarizing the schemas described in this study, which seem to be used by speakers in the creation of new words. All of them result in the formation of blended words ending in ‘-gasm’. 106 Some blends show a relationship of cause and effect between their source words. The words that behave in this way are instantiations of the following schemas: (1) <[[X]Ni gasm]Nj ↔ [A sensation of physical or bodily pleasure produced by SEMi]j> (2) <[[X]Ni gasm]Nj ↔ [A feeling of excitement or enthusiasm produced by SEMi]j> 107 The main difference between the schemas in (1) and (2) is the precise meaning of the resulting feeling triggered by the cause. While this is a feeling of physical of bodily pleasure in the first case, it is actually a feeling of excitement or enthusiasm in the second. An example of a word created by means of the schema in (1) is toygasm. The word neologasm, on the other hand, is an instantiation of the schema in (2). 108 In addition, lower-level schemas have been identified for both cases. They are represented as follows: (1.1.) <[[x]Ni gasm]Nj ↔ [A sensation of physical or bodily pleasure produced by eating or drinking SEMi]j> (2.1.) <[[X]Ni gasm]Nj ↔ [A feeling of excitement or enthusiasm produced by listening to music by / the sound of SEMi]j> 109 Each of the subschemas presented above differ from their general schemas in the degree of semantic specificity of the causing entity. In (1.1.), the cause resulting in a pleasurable effect is the activity of eating or drinking something, as can be seen in beefgasm, while in (2.1.), the cause is differently specified, and is restricted to cases where the feeling of excitement is produced by the activity of listening to a specific type of music or sound, as in guitargasm. 110 In other blends, like clowngasm, the first participant fulfils the semantic role of an experiencer, that is, the first participant experiences the feeling of pleasure. The constructional schema accounting for these words is the following: (3) <[[X]Ni gasm]Nj ↔ [A feeling of pleasure or excitement experienced by SEMi]j> 111 Two subschemas can be found within the group of experiencer blends, whose main difference is the nature of the first source word. As represented below, the first source word is a proper noun in (3.1.), as in Horgasm, while it is a common noun in (3.2.), as in geekgasm. Lexis, 14 | 2019 116 (3.1.) <[[X]PNi gasm]Nj ↔ [A feeling of pleasure or excitement experienced by SEMi]j> (3.2.) <[[X]CNi gasm]Nj ↔ [A feeling of pleasure or excitement experienced by SEMi]j> 112 Another schema for the creation of words in ‘-gasm’ is the one represented in (4). This schema is responsible for the formation of coordination blends like crygasm, that is, blended words whose source words stand in a relationship of coordination. (4) <[[X]Ni gasm]Nj ↔ [A feeling of pleasure or excitement experienced together with (a feeling of) SEMi]j> 113 The schema represented in (5) is different from the rest in that the first source word is an adjective. The use of this schema results in adjectival blends in which the first element premodifies the feeling denoted by the second element. One instance of these blends is fakegasm. (5) <[[X]Ai gasm]Nj ↔ [An orgasm of the nature described by SEMi]j> 5. Conclusion 114 The present study has proved that the splinter ‘-gasm’ is very productive. It is included in a huge list of words and many of these words are cases of hapax legomena in the corpus used, that is, words that appear with the minimum possible frequency in a corpus. This is an indicator that novel words containing ‘-gasm’ are created regularly by speakers, who seem to use the sequence quite naturally expecting other language users to recognize and understand their coinages. 115 The semantic analysis of the relationships between the blends’ source words has allowed for the classification of blends in different categories. In this way, I have distinguished blends that are based on a relationship of causality from the rest. Within this group, two situations are found: in some of them, the splinter seems to maintain the meaning of the original word it comes from. In others, however, the meaning of the splinter presents slight variations from the original one. Another group of blends is characterized by a first source word with the semantic role of experiencer, an entity that is affected by a feeling of pleasure or excitement, denoted by the sequence ‘-gasm’. In addition, some blends seem to have been formed by means of coordination. In these words, we cannot identify a semantic head, as both source words seem to be equally important. Furthermore, these words denote a feeling of pleasure or excitement accompanied by another feeling. Blends formed by an adjective followed by ‘-gasm’ seem to form another category, as their semantic behaviour is constant, with the adjectives in all the words acting as premodifiers of the meaning of ‘-gasm’ (pleasure or excitement). Some other words could be considered to form other minor groups or be a part of the previously described ones, but the corpus does not contain sufficient data to set up new groups. 116 As opposed to purely concatenative approaches to morphology, Construction Morphology has proved to be an appropriate tool to account for these formations. The use of schemas and subschemas appears to be very useful to explain the formation of the blends in the corpus, as well as the differences between them. 117 Future research should attempt to find more differentiated groups of blends according to the nature of their source words. Moreover, they should study other cases of blend Lexis, 14 | 2019 117 splinters to check whether the same patterns that have been described in this work can be identified in blends containing other sequences. BIBLIOGRAPHY ADAMS Valerie, 1973, Introduction to Modern English Word-Formation, London: Longman. ALGEO John, 1977, “Blends, a structural and systematic view”, American Speech 52, Durham: Duke University Press, 47-64. ARDNT-LAPPE Sabine, 2014, “Analogy in suffix rivalry: the case of English –ity and -ness”, English Language and Linguistics, 18(3), Cambridge: Cambridge University Press, 537-568. ARDNT-LAPPE Sabine & PLAG Ingo, 2013, “The role of prosodic structure in the formation of English blends”, English Language and Linguistics, 17(3), Cambridge: Cambridge University Press, 537-563. BAUER Laurie, 1983, English word-formation, Cambridge: Cambridge University Press. BAUER Laurie, 2001, Morphological Productivity, Cambridge: Cambridge University Press. BAUER Laurie, 2006, “Compounds and minor word-Formation types”, in AARTS Bas & MCMAHON April (Eds.), The Handbook of English Linguistics, Hoboken, New Jersey: Blackwell, 483-506. BAUER Laurie, LIEBER Rochelle & PLAG Ingo, 2013, The Oxford Reference Guide to English Morphology, Oxford: Oxford University Press. BECKER Thomas, 1993, “Back-formation, cross-formation and ‘bracketing paradoxes’”, in BOOIJ Geert & VAN MARLE Jaap (Eds.), Yearbook of Morphology, 1993, Dordrecht: Springer, 1-25. BELIAEVA Natalia, 2014, “A study of English blends: From structure to meaning and back again”, in PASTER Mary, RAINER Franz, SPENCER Andrew & STUMP Greg (Eds.), Word Structure, 7, Edinburgh: Edinburgh University Press, 29-54. BERMAN J.M., 1961, “Contribution on Blending”, Zeitschrift für Anglistik und Amerikanistik, 9, Germany: Walter De Gruyter, 278-281. BOOIJ Geert, 2010, Construction Morphology, Oxford: Oxford University Press. BOOIJ Geert, 2015, “Construction Morphology”, in HIPPISLEY Andrew & STUMP Gregory (Eds.), The Cambridge Handbook of Morphology, Cambridge: Cambridge University Press, 424-448. BURZIO Luigi, 1998, “Multiple correspondence”, Lingua, 104(1), Amsterdam: Elsevier, 79-109. CANNON Garland, 1986, “Blends in English word formation”, Linguistics, 24, Cambridge: Cambridge University Press, 725-753. CANNON Garland, 2000, “Blending”, in BOOIJ Geert, LEHMANN Christian & MUGDAN Joachim (Eds.), Morphology: An International Handbook on Inflection and Word-Formation, Berlin and New York: Walter de Gruyter, 952-956. CHOROLEEVA Kornelia, 2015, “Morpho-semantic groups of lexical blends”, Scientific Works Of University Of Food Technologies, 62, 902-904. Lexis, 14 | 2019 118 CORREIA SAAVEDRA David, 2014, “Automatically identifying blend splinters that are morpheme candidates”, Digital Scholarship in the Humanities, 31(1), Oxford: Oxford University Press, 55-71. CRYSTAL David, 1995, The Cambridge Encyclopaedia of the English Language, Cambridge: Cambridge University Press. FANDRYCH Ingrid, 2008, “Sub-morphemic elements in the formation of acronyms, blends and clippings”, Lexis, 2, 103-121. FAUCONNIER Gilles & TURNER Mark, 1998, “Conceptual integration networks”, Cognitive Science, 22(2), 133-187. FAUCONNIER Gilles & TURNER Mark, 2002, The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities, New York: Basic Books. FISCHER Roswitha, 1998, Lexical Change in Present Day English: A Corpus Based Study of the Motivation, Institutionalization, and Productivity of Creative Neologisms, Tübingen: Gunter Narr Verlag. GEERAERTS Dirk & CUYCKENS Hubert, 2007, The Oxford Handbook of Cognitive Linguistics, Oxford: Oxford University Press. GOLDBERG Adele E., 1995, Constructions: A Construction Grammar approach to argument structure, Chicago and London: Chicago University Press. GOLDBERG Adele E., 1998, “Patterns of experience in patterns of language”, in TOMASELLO Michael (Ed.), The New Psychology of Language (Cognitive and Functional Approaches to Language Structure), Vol. 1, Mahwah, N.J.: Lawrence Erlbaum, 203-219. GOLDBERG Adele E., 1998, “Construction grammar”, in NADEL Lynn (Ed.), Encyclopedia of cognitive science, London: MacMillan. GOLDBERG Adele E., 1995, Constructions at Work: The Nature of Generalization in Language, Oxford: Oxford University Press. GRIES Stephan Th., 2004a, “Isn’t that Fantabulous? How Similarity Motivates Intentional Morphological Blends in English”, in ACHARD Michel & KEMMER Suzanne (Eds.), Language, Culture, and Mind, Stanford: CSLI Publications, 415-428. GRIES Stepha Th., 2004b, “Shouldn’t it be breakfunch? A quantitative analysis of blend structure in English”, Linguistics, 42, Cambridge: Cambridge University Press, 639-667. HILPERT Martin, 2014, Construction Grammar and its application to English, Edinburgh: Edinburgh University Press. KELLY Michael H., 1998, “To “brunch” or to “brench”: some aspects of blend structure”, Linguistics, 36(3), Cambridge: Cambridge University Press, 579-590. KEMMER Suzanne, 2006, “Schemas and lexical blends”, in CUYCKENS Hubert, BERG Thomas, DIRVEN René & PANTHER Klaus-Uwe (Eds.), Motivation in Language: From Case Grammar to Cognitive Linguistics. A Festschrift for Günter Radden, Amsterdam and Philadelphia: John Benjamins, 69-97. LALIC-KRSTIN Gordana, 2014, “Forms and meanings of the source word Armageddon in English lexical blends”, in PRĆIĆ Tvrtko et al. (Eds.), Engleski jezik i anglofone književnosti u teoriji i praksi: Zbornik u čast Draginji Pervaz, Filozofski fakultet: Novi Sad, 257-273. LEHRER Adrienne, 1996, “Identifying and interpreting blends: An experimental approach”, Cognitive Linguistics, 7(4), Berlin and New York: Walter de Gruyter, 359-390. Lexis, 14 | 2019 119 LEHRER Adrienne, 1998, “Scapes, holics and thons: The semantics of English combining forms”, American Speech, 73(1), Durham, N.C.: Duke University Press, 3-28. LEHRER Adrienne, 2007, “Blendalicious”, in MUNAT Judith (Ed.), Lexical Creativity, Texts and Contexts, Amsterdam and Philadelphia: John Benjamins, 115-133. LÓPEZ-RÚA Paula, 2002, “On the Structure of Acronyms and Neighbouring Categories: a Prototype- based Account”, English Language and Linguistics, 6(1), Cambridge: Cambridge University Press, 31-60. PLAG Ingo, 2003, Word-Formation in English, Cambridge: Cambridge University Press. PLAG Ingo & BAAYEN Harald, 2009, “Suffix ordering and morphological processing”, in CARLSTON Greg & STUMP Gregory (Eds.), Language, 85(1), New York: Linguistic Society of America, 109-152. POND Louise, 1914, Blends: Their Relation to English Word Formation, Heidelberg: Winter. RADDEN Gunter & DIRVEN René, 2007, Cognitive English Grammar, Amsterdam and Philadelphia: John Benjamins. RAFFELSIEFEN Renate, 2004, “Absolute ill-formedness and other morphophonological effects”, Phonology, 21, Cambridge: Cambridge University Press, 91-142. SCHREUDER Robert & BAAYEN Harald, 1997, “How complex simple words can be”, Journal of Memory and Language, 37, Amsterdam: Elsevier, 118-139. SOUDANEK L.J., 1998, “The Relation of Blending to English Word-Formation: Theory, Structure, and Typological Attempts”, in DRESSLER Wolfgang & MEID Wolfgang (Eds.), Proceedings of the Twelfth International Congress of Linguists: Vienna, August 28-September 2, 1977, Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, 462-466. STERIADE Donca, 2000, “Paradigm uniformity and the phonetics-phonology boundary”, in BROE M. & PIERREHUMBERT J. (Eds.), Papers in laboratory phonology V: Acquisition and the lexicon, Cambridge: Cambridge University Press, 313-334. TAYLOR John R., 2012, The Mental Corpus: How Language is Represented in the Mind, Oxford: Oxford University Press. UNGERER Friedrich & SCHMID Hans-Jörg, 1996, An Introduction to Cognitive Linguistics, London: Longman. APPENDIXES Appendix 1 Table 1: Frequency of the words ending in –gasm in the corpora FREQUENCY Word En TenTen15 GlowBe COCA orgasm 39,845 5,966 1,744 Afrigasm 4 6 Lexis, 14 | 2019 120 Allahgasm 1 Amerigasm 1 Artgasm 8 Asianporngasm 1 Assgasm 5 Astrogasm 1 Audiogasm 7 Aurgasm 419 Bassgasm 10 Bedgasm 1 Beefgasm 1 Beergasm 1 Berbagasm 3 Berlinergasm 5 Biebergasm 1 1 Blissgasm 1 Blogasm 7 Bookgasm 10 5 Bootygasm 1 Braingasm 4 Breathgasm 2 Cakegasm 2 Cameragasm 1 Cargasm 1 2 Carnivorgasm 1 Chartgasm 1 Chillgasm 1 Lexis, 14 | 2019 121 Chocgasm 2 Chocogasm 4 Choirgasm 1 Christgasm 1 Clowngasm 2 Colbertgasm 1 Coregasm 4 1 Crabgasm 1 Crygasm 3 Cutegasm 1 Cyborgasm 6 Dancegasm 1 Deathgasm 94 Debtgasm 1 Deli-gasm 1 Delish-o-gasm 1 Designgasm 2 Drawergasm 1 Dreamgasm 1 Eargasm 122 17 Electrogasm 1 Exploragasm 1 Explorgasm 2 Extravagasm 2 Eyegasm 17 1 Factgasm 3 Failgasm 1 Lexis, 14 | 2019 122 Fangasm 65 3 1 Fashiongasm 1 Feargasm 1 Femaleorgasm 18 Femgasm 6 Femmegasm 3 Flavourgasm 1 Floorgasm 2 Foodgasm 114 14 Footgasm 1 Forevergasm 1 Fruitgasm 1 2 Gamegasm 1 Gamingasm 1 Gastronogasm 1 Geekgasm 37 4 1 Goalgasm 2 1 Googasm 2 Goregasm 7 2 Gorgasm 11 Guitargasm 1 Heartgasm 4 Hentaigasm 2 Herogasm 8 Horrorgasm 2 Horgasm 2 Hortgasm 2 Lexis, 14 | 2019 123 Ideagasm 1 Infogasm 2 Internetgasm 1 Joygasm 12 4 Ladygasm 86 Lashgasm 5 Leathergasm 1 Lesbigasm 1 Lightning-gasm 1 Literarygasm 1 Luxogasm 1 Macgasm 61 9 Mapgasm 1 Mangasm 60 Meatgasm 6 Metalgasm 1 Mindgasm 6 1 Moneygasm 1 Mooregasm 6 Moregasm 43 4 Morgasm 23 1 Mouthgasm 12 3 Mustangasm 1 Neologasm 2 Nerdgasm 139 30 1 Newsgasm 5 Nipplegasm 1 Lexis, 14 | 2019 124 Noisegasm 11 Obamagasm 17 2 Obamaorgasm 1 Operagasm 8 Orbgasm 1 Orchidgasm 1 Oreogasm 2 Orgygasm 1 Oxtailgasm 1 Papergasm 3 Photogasm 2 Pinkgasm 1 Plankgasm 1 Plantgasm 1 Polishgasm 1 Popgasm 4 Porkgasm 1 Porngasm 17 Prayergasm 3 Prorgasm 2 Pyrogasm 1 Queergasm 1 Ragegasm 2 Releasegasm 1 Rockgasm 1 Scenerygasm 1 Schizorgasm 1 Lexis, 14 | 2019 125 Scoregasm 3 Sexgasm 1 Shoegasm 1 Showgasm 7 Skygasm 1 Snoorgasm 1 Snoregasm 1 Soapgasm 1 Soulgasm 18 Soundgasm 1156 Sportsgasm 1 Squeegasm 6 2 Storygasm 7 Studygasm 1 Stylegasm 1 Suregasm 1 Tabloidgasm 1 Tastegasm 1 Thoughtgasm 1 Tintingasm 2 Torygam 7 Tourgasm 11 2 1 Toygasm 1 Trigasm 1 TVgasm 2 1 Twittergasm 1 Vampiregasm 1 Lexis, 14 | 2019 126 Vetogasm 2 Vowelgasm 1 Wargasm 32 2 4 Whoregasm 4 Winegasm 1 Wordgasm 1 Yawgasm 1 Yawngasm 7 Yogasm 1 Appendix 2 Table 2: Cause and effect blends where ‘-gasm’ means ‘physical or bodily pleasure’ Word Definition Example “Asianporngasm - extremely hot resource An orgasm reached by watching Asian with a lot of beautiful and sexual Asian Asianporngasm porn. chicks” (www.nakedsexblogs.net). Taken from EnTenTen15. “Hayley enjoyed her anal adventure especially since she was able to experience A feeling of physical pleasure reached Assgasm multiple through the stimulation of the anus. assgasms” (www.mobilecarriers.org). Taken from EnTenTen15. “We spoke to one mum who, without A feeling of physical pleasure realising what it was, had a birthgasm Birthgasm experienced while giving birth to a while giving birth to her second child, and child. never told anyone about it.” (www.netmums.com). “My god. That chillgasm was so worth the “An orgasm as a result of being really, Chillgasm money we spent on all that Indo” really relaxed” (UrbanDictionary). (UrbanDictionary) “An orgasm achieved while ““coregasm,” named thusly because Coregasm performing a core exercise during a abdominal exercises tend to spark the work-out” (UrbanDictionary). sensation.” (www.healthland.time.com). “When a fail feels good” “I had a failgasm yesterday” Failgasm (UrbanDictionary) (UrbanDictionary). Lexis, 14 | 2019 127 “When a sudden event like the sighting of a cop on the highway causes electric-like jolts of fear to “Why are you hyper-ventilating?” "iPhone Feargasm shoot throughout your body. feargasm! I can’t find it anywhere!” Reminiscent of an orgasm but not as (UrbanDictionary) fun, especially if followed by speeding ticket.” (UrbanDictionary) “One of the diners at an adjacent table had a noisy foodgasm over this dessert, so we The feeling of pleasure one gets when Foodgasm tried it on our second visit” eating something. (www.magoguide.net). Taken from EnTenTen15. An orgasmic feeling obtained when “Megan footgasms when I rub her feet nice Footgasm getting a very good foot massage. and hard” (UrbanDictionary). An orgasm reached by watching “Raw Hentaigasm - Stream Hentai” Hentaigasm hentai, that is, animated, cartoonish (www.mirvramke.by). Taken from porn. EnTenTen15. When a good full-bodied laugh can “You can even get a laughgasm from certainly help you experience the laughing so hard” Laughgasm relief, release and feeling of (www.neologisms.rice.edu). Taken from refreshment you might have from an EnTenTen15. orgasm” (www.mirthmaven.wordpress.com). “We tried all kinds of fair fried things and A feeling of pleasure obtained by we had a mouthgasm when we had fried Mouthgasm eating something one loves. oreos.” (www.addymeira.net). Taken from EnTenTen15. “Some women can orgasm just through stimulation of breasts and nipples. No, An orgasm resulting from the Nipplegasm nipplegasms aren’t a myth!” stimulation of the nipples. (www.girlsandbabes.in). Taken from EnTenTen15. “Haired bosses will have an An orgasm resulting from taking part Orgygasm orgygasm” (www.slashdot.org). Taken from in an orgy. EnTenTen15. An orgasm experimented while “Porngasm is waiting for you to enjoy free Porngasm watching porn. adult video chat” (www.chaturbate.com). “Do you want sexgasms, playgasms, orgasms and laughgasms? Then this is the An orgasm reached when talking about perfect raucous, information-packed, fun, Sexgasm or dealing with sex. play shop for you” (www.pamelamadsen.org). Taken from EnTenTen15. Lexis, 14 | 2019 128 An orgasm experienced when using “Toygasm! “The Insider’s Guide to Sex Toys Toygasm sex toys for stimulation. and Techniques” (COCA). “An almost sexually blissful feeling experienced via practicing yoga that is “When that 90 minutes of hot horror was Yogasm so physically and spiritually fulfilling over, I had the biggest yogasm ever!!!” that it is akin to an orgasm” (UrbanDictionary) (UrbanDictionary) Appendix 3 Table 3: Cause and effect blends modelled on ‘foodgasm’ Word Definition Example “The Wagyu beef just melts in my mouth whilst the Pleasure obtained by eating grain fed beef is so tender and juicy. Beefgasm beef. #Beefgasm” (www.annna.net). Taken from EnTenTen15. “I also love beer and will be hosting the Beergasm on Pleasure obtained by drinking Thursday afternoon!” Beergasm beer. (www.porcfest2015.sched.org). Taken from EnTenTen15. “A mind blowing double “I nearly had a cakegasm at the table” Cakegasm orgasm induced by cakes.” (www.goodreads.com). (UrbanDictionary) “it was nice to know that I could create such a pleasurable working environment by just bringing Chocgasm, A feeling of pleasure derived in a five dollar cake... and the chocgasms that Chocogasm from eating chocolate. followed were only a bonus” (www.straightwhiteguy.mu.nu). Taken from EnTenTen15. Pleasure obtained by eating a “Walk in, get a table pretty easily, and be prepared Crabgasm crustaceous. to have a crabgasm” (www.ohjoy.blogs.com). “Deligasm can be eating great pastrami, corned beef Deligasm sandwiches at a good Jewish, kosher or other kosher-style deli restaurant.” (UrbanDictionary) Delish-o- Pleasure obtained by eating gasm something delicious. Lexis, 14 | 2019 129 “Used to liken the pleasure of eating flavorful food to the “That steak was so juicy and tender that I think I euphoria experienced during Flavourgasm just had a flavorgasm in my mouth” sexual release” (www.neologisms.rice.edu). (www.neologisms.rice.edu). Taken from EnTenTen15. Pleasure derived from eating a “Judy just had a Fruitgasm from that piece of Fruitgasm piece of fruit. watermelon!” (UrbanDictionary) “We were there for the cheeseburger meatgasm... Pleasure obtained by eating Meatgasm Seriously, you’ve never seen such joy” meat. (www.evilbastard.net). Taken from EnTenTen15. “Lisa is also a contributor to Funny Times newspaper where her parody of Oreo cookies was recently featured alongside humor greats Dave Pleasure obtained by eating Oreogasm Barry and Garrison Keillor--an experience she Oreo biscuits. found so stimulating, she had an Oreogasm” (www.myfunnybooks.biz). Taken from EnTenTen15. “It’s the oxtail soup. Pleasure derived from eating Oxtailgasm Oxtailgasm” (www.kualalumpur-travel.info). Taken something containing oxtail. from EnTenTen15. “Just because I have this recipe listed under Pleasure obtained by eating Porkgasm Porkgasm in my recipe file doesn’t mean it’s a pork. fetish” (www.bbqaddicts.com). Taken from GloWbE. “Sorry to bombard you with sloppy foods. I know the last post was pretty much a soupgasm. You may Pleasure obtained by drinking Soupgasm be worried this blog is going to become reduced to a soup. soup blog” (www.ckrecipes.com). Taken from GloWbE. “A state of ecstasy achieved during the consumption of “I would be up for Swirll; I haven’t had a Swirllgasm Swirllgasm Swirll frozen yogurt similar to in a couple of weeks” (www.neologisms.rice.edu). that achieved during intercourse” (www.neologisms.rice.edu). “The other tastegasm of 2011 came at Metamorfosi with Chef Roy Caceres’ glazed eel paired with Pleasure obtained by eating Tastegasm ground spelt and a punchy carpione onion and something tasteful or delicious. vinegar sorbet” (www.parlafood.com). Taken from EnTenTen15. “We were well into a winegasm when the second Pleasure obtained by drinking Winegasm course arrived” (www.magoguide.net). Taken from wine. EnTenTen15. Lexis, 14 | 2019 130 Appendix 4 Table 4: Cause and effect blends where ‘-gasm’ means ‘a feeling of excitement or enthusiasm’ Word Definition Example A feeling of enthusiasm “Afrigasms feel as good as they do because they confirm experienced when dealing with Afrigasm the way media . . . taught us to view the continent” or watching images from (www.vijana.fm). Taken from EnTenTen15. Africa. “Having intense joy or passion “Listening to Toby Keith’s song “Courtesy of the Red, Amerigasm upon extreme patriotism for White, & Blue (The Angry American)” gave me an the U.S” (UrbanDictionary). Amerigasm” (UrbanDictionary) “Astrogasm is about being and having an explosion of A feeling of excitement related Astrogasm excitement about the stars, cosmos, and life” to the stars and the universe. (www.astrogasm.com). Taken from EnTenTen15. “Bedgasm (n): a feeling of “It was a three hour drive in the middle of the night, I euphoria experienced when could barely stay awake. When I got home, I climbed Bedgasm climbing into bed” under the covers and had a ten-minute (www.architecturendesign.net). bedgasm” (UrbanDictionary) Taken from EnTenTen15. The feeling of euphoria or “Again it’s all hearsay as we can’t change the past but excitement derived from I’m pretty sure my timeline would all have a Berbagasm Berbagasm watching football player if he was to help bring home the title for us.” Dimitar Berbatov play a good (www.thefaithfulmufc.com). Taken from GloWbE. match. “I don’t know the right word for it, so I’m coining “Berlinergasms”. I was on the tram recently and A feeling of euphoria of a overheard an English guy turning to his two friends Berlinergasm Berliner due to their place of and saying loudly “I fucking love living in Berlin. I just residency. love it. It’s just so fucking great”” (www.uberlin.co.uk). Taken from GloWbE. “The feeling you get when you “Dude, last night I wrote a new blog post about that check your blog stats and you crazy new action movie. This morning when I checked Blogasm see way more visitors than you my stats and saw I had over 2,000 visits, I had a serious expected” (UrbanDictionary). blogasm!” (UrbanDictionary). Feeling of enthusiasm “Bookgasm – reading material to get excited about” Bookgasm experienced when reading a (www.netfacet.net). Taken from EnTenTen15. book. Feeling of enthusiasm “If you really want to have a cameragasm have him take Cameragasm experienced when using a you up to the pro floor and take a look at the Hassys” camera to take pictures. (www.shutterstock.com). Taken from GloWbE. Lexis, 14 | 2019 131 “I know the car isn’t “beautiful” by Ferrari standards, Feeling of excitement or but it still satisfies every cargasm and luxogasm need Cargasm enthusiasm when seeing or any of us may ever have” (www.hotelroomsdirect.uk). owning a car. Taken from EnTenTen15. Feeling of enthusiasm “Prepare for chart-gasm and graph-overload. And Chartgasm experienced when finding or prepare to probably say, ‘Grr’” (www.fazed.net). Taken using charts. from EnTenTen15. “Most likely the expression on their faces as they pore over their “proofs” is the same as a Christian fundy in Excitement derived from the Christgasm the throes of a religious belief in Jesus Christ. Christgasm” (www.blogs.discovermagazine.com). Taken from GloWbE. “The reaction one feels when being exposed to something overly cute. this may be an “When Holly saw the baby trying to dance, she had a Cutegasm emotional, physical or even cutegasm” (UrbanDictionary). sexual response” (UrbanDictionary). “A strong feeling of excitement brought on by anything dark or “That necrophile creep had a deathgasm in the morgue Deathgasm of death, such as music and art looking at all those bodies” (UrbanDictionary). felt by death enthusiasts” (UrbanDictionary). A feeling of enthusiasm derived “And what architecture-porn would be complete Designgasm from architecture and the without a designgasm” (www.nolandgrab.org). Taken activity of designing. from EnTenTen15. “They were tricky (my first drawers), but once they A feeling of excitement or were finished they slid like butter, and latched so enthusiasm experienced when Drawergasm sweetly (Laura Goldhamer called opening and closing opening and closing drawers them a “drawergasm”)” (www.freeteaparty.org). Taken that slide perfectly. from EnTenTen15. A feeling of maximal “I nearly had an eargasm while listening to his enthusiasm or excitement Eargasm performance of Rachmaninoff’s Second Piano experienced when listening to Concerto” (UrbanDictionary). music. A feeling of excitement or “New York’s Times Square has always been an eyegasm enthusiasm when seeing Eyegasm of advertising” (www.ringofstars.ru). Taken from something overwhelmingly EnTenTen15. beautiful. Lexis, 14 | 2019 132 “Welcome to the first edition of Calais Migrant Factgasm, in which I quite metaphorically round up A feeling of excitement when Factgasm every piece of internet about the Calais migrants and discovering new facts. incarcerate it in the detention centre of my blog” (www.davidcharles.info). Taken from EnTenTen15. “Brix’s ultimate test for a product is the “fashiongasm”. That is, if something fills her with an explosive, must- A feeling of excitement have sense of need then she knows she’s onto a hit. FYI, experienced when watching Brix’s recent fashiongasms have come from NEWGEN Fashiongasm fashion items, such as clothes shoe designer Sophia Webster and the leather tote bags or shoes. of Sophie Hulme” (www.fashioneditoratlarge.blogspot.com). Taken from GloWbE. “Ballroom addict 1: Say, did you watch that PBS A feeling of excitement America’s Ballroom Challenge? Floorgasm experienced when dancing on a dancefloor. Obsessed Ballroom Addict: OMG did you SEE Ben and Shalene? Floorgasm!” (UrbanDictionary) A feeling of excitement of “When playing Command and Conquer 3 it sounded enthusiasm given by the visual Gamegasm and looked so good I had a components, aesthetics, or gamegasm” (UrbanDictionary). contents of a game. A feeling of excitement given “Easily the most deluxe gaming product I’ve ever seen! Gamingasm by the activity of playing a #gamingasm” (Twitter). game “Bench! Ball! Goal! Arjen Robben watched the ball from A feeling of excitement the bench for 54 minutes. He was subbed on. And Goalgasm obtained when one’s favourite celebrated his personal goal-gasm with his first touch of team scores a goal. the ball after 37 seconds” (www.espnfc.us). Taken from EnTenTen15. “Tech media starting to catch on that nothing Google can make will spell the end for Microsoft. Some writers A feeling of enthusiasm by the Googasm fake a Googasm and manage to fool an inexperienced success of Google. few readers, but those of us in the know can spot the act” (www.theregister.co.uk). Taken from GloWbE. “The word Goregasm was invented to describe the A feeling of enthusiasm feelings an Al Gore lover gets when doing anything Al Goregasm experienced when doing Gore related, an event referred to as a Goregy when anything related to Al Gore. occurring collectively” (www.realcty.org). Taken from EnTenTen15. Lexis, 14 | 2019 133 Name of a band, presumably denoting a feeling of Gorgasm excitement derived from gore- related, that is, sanguine- related things. Probably a feeling of excitement when facing Horrorgasm horror-related things or activities. Taking ‘horticulture’ as its first “This is truly one of the most elegant plants in the source word, this word stands genus arum, and to catch them in perfect flower was a for the feeling of enthusiasm of Hortgasm true “hortgasm” moment” excitement experienced when (www.juniperlevelbotanicgarden.org). Taken from seeing beautiful plants and EnTenTen15. flowers. A feeling of excitement “Holy cow that is such a good idea!!! I think I just had Ideagasm produced by having a an ideagasm” (UrbanDictionary). wonderful idea. A feeling of excitement “Why does it feel so good just to answer and delete Internetgasm produced by surfing the emails? #internetgasm” (Twitter). Internet. Name of a product for beauty aimed at getting perfect eyelashes. Presumably standing Lashgasm for a feeling of enthusiasm produced by having perfect eyelashes. The feeling one gets when “I’m intoxicated by the smell of real leather and Leathergasm admiring a beautiful leather fascinated by the rawness of handmade leather goods. work. #leathergasm” (Twitter). Excitement or enthusiasm Lightning- resulting from a good use of gasm lightning techniques in a film. “I know the car isn’t “beautiful” by Ferrari standards, A feeling of enthusiasm but it still satisfies every cargasm and luxogasm need Luxogasm produced by seeing or owning any of us may ever have” (www.hotelroomsdirect.uk). luxury items. Taken from EnTenTen15. A feeling of excitement when Macgasm using or owning a Mac product designed by Apple. Lexis, 14 | 2019 134 “This is post number six in the ongoing #mapgasm A feeling of enthusiasm series of posts on maps found on the interwebs that I Mapgasm produced by looking at maps. like. Yes, it’s another map. Yes, it’s another Tube map” (www.vicchi.org). Taken from EnTenTen15. A feeling of excitement “The Super Bowl is an enormous money-gasm, but it is Moneygasm produced by receiving a so far more for the TV people than the NFL” quantity of money. (www.sportschump.net). Taken from EnTenTen15. “On the next row over was the exclusive Mustang Club. A feeling of enthusiasm More Mustangs than you could ever want or dream Mustangasm produced by admiring a about were all lined up for your personal Mustang car. Mustangasm” (www.barry-stein.net). Taken from GloWbE. “Volunteers submit over 1000 new definitions to Excitement from coining or Neologasm Urbandictionary every day. It ain’t fo the money; it’s all finding a neologism. about the neologasms” (UrbanDictionary). “Put simply, if there was even a remote possibility of us A feeling of excitement being turned into a colony the Daily Mail would have a Newsgasm experienced when having or newsgasm and put it all over pages 1-27” displaying news. (www.tboverse.us). Taken from EnTenTen15. Name of an electronic music Noisegasm duo. Probably the feeling one gets when hearing loud noises. A feeling of excitement or “If your so happy paying subsidies, then your gonna enthusiasm related to ex- Obamagasm, have an ObamaGasm when you get your bill for the President Barack Obama or any Obamaorgasm Socialist Medical Program (National Health Care)” policies issued by his (www.fullertonsfuture.org). Taken from EnTenTen15. government. In the context provided, a “Newsweek promised to demystify the viral orbgasm in feeling of excitement people an article published Sunday called “Why Donald Trump experience when talking about Touched a Glowing Orb in Saudi Arabia”. The article Orbgasm Donald Trump touching a cooly explains that the image depicts Trump at the glowing orb in a visit to Saudi opening of the Global Center for Combatting Extremist Arabia. Ideology in Riyadh” (www.theslot.jezebel.com). “The challenge: have fun in Singapore ... the mist A feeling of excitement house, the cool house, and the VIP Orchid Garden. I Orchidgasm produced by seeing beautiful think I had an Orchid flowers. orchidgasm” (www.greenteafrappuccino.loseweightquickly.space). Taken from EnTenTen15. A feeling of excitement experienced when looking at “Ultimate want. I just had a papergasm here” Papergasm stationary material or items (Pinterest). made of paper. Lexis, 14 | 2019 135 Feeling of excitement related to Pinkgasm the use of the colour pink. “Another plankgasm! Yay! It seems I am now just Feeling of excitement getting on with the planks. If you are looking for more experienced when being Plankgasm challenging plank variations, check this out” capable of doing plank (www.thepeakconditionproject.com). Taken from abdominal exercises. GloWbE. “You can read some about this at Plantgasm, if you’re A feeling of enthusiasm when interested. Passiflora caerulea and maybe other Plantgasm admiring beautiful plants or Passiflora spp. (passion flower)” flowers. (www.plantsarethestrangestpeople.blogspot.com). Taken from GloWbE. “if Michael Bay put out a movie called “Sh*t Blows Up” and made two hours of a seat-shaking, AC/DC blaring, A feeling of enthusiasm or mushroom-clouded pyrogasm of splodey bits, in super- Pyrogasm excitement experienced when mega 4D (enveloping you in a smog smelling of burnt admiring pyrotechnics. rubber and man sweat) it would be the biggest selling movie of all time” (www.tor.com). Taken from GloWbE. “Lubuntu , Mythbuntu , Edubuntu , Kubuntu , and A feeling of enthusiasm Ubuntu Studio , have all been updated, absorbing and produced by the commercial expanding upon various features present in Ubuntu Releasegasm release of a new product (in the 11.04. Absolute 13.2.2, a light-weight version of example provided, the product Slackware, rounds out the releasegasm of the last 24 being a software). hours” (www.matias.root.zone). Taken from EnTenTen15. The feeling of excitement one gets when admiring the “Can you say scenerygasm: The Princess and the Frog” Scenerygasm beautiful scenery of a film or a (www.derynsharp.tk). Taken from EnTenTen15. work of art. “Palace can thank Hugh O’Higgins for a 90 th minute 45 A feeling of excitement yard piledriver to seal victory while Scoregasms will be obtained when a footballer of grateful to Ladzio player Ronald Carroll who found Scoregasm one’s favourite team scores a himself taking an early bath after picking up two goal. yellow cards, both for dissent towards the referee” (www.collegetribune.ie). Taken from EnTenTen15. A feeling of excitement when “When I saw those pink Mary Janes, I had the most Shoegasm seeing a pair of beautiful shoes. amazing shoegasm of my life” (UrbanDictionary). A feeling of enthusiasm “Have you seen the new episode of Doctor Who? man Showgasm produced by attending or that show gives me the best watching a wonderful show. showgasm” (UrbanDictionary). A feeling of excitement “Elly had a soundgasm every time she heard a cat Soundgasm experienced when hearing meow” (UrbanDictionary) something. Lexis, 14 | 2019 136 “I’ve been reserved so far, but I feel a sports-gasm A feeling of enthusiasm or coming on that won’t subside until the end of the NFL Sportsgasm excitement when playing or season” (www.andrewdupont.net). Taken from watching sports. EnTenTen15. “Join Storygasm, the queer storytelling and roleplaying Enthusiasm derived from Storygasm collective, to tell collaborative stories!” reading or listening to a story. (www.nineworlds.co.uk). “Directed by Pierre Koralnik and starring Anna Karina, A feeling of excitement the movie might best be described as an 87-minute Stylegasm produced by the style of a film, stylegasm” (www.thesamecinemaeverynight.net). Taken a person or another object. from EnTenTen15. A feeling of enthusiasm Tabloidgasm produced by reading an article in a tabloid. A feeling of excitement when “I believe I just experienced a thoughtgasm from staring Thoughtgasm having or coming up with new at Mrs. Smith’s ass!” (UrbanDictionary). thoughts and ideas. “Oh, and yes -Spielberg needs to make that Enthuasiasm related to the Indianapolis movie! Maybe after he gets this Tintingasm Tintingasm Tintin character or film. out of his system he can make another actual movie” (www.aintitcool.com). Taken from GloWbE. A feeling of excitement of a musician when going on tour Tourgasm or a fan when going to concerts. A feeling of excitement when “Greys Anatomy, Scandal & How to get away with TVgasm watching material one loves on murder!!! #tvgasm” (Twitter). TV. “People on social media seemed to have a simultaneous A feeling of excitement when Twittergasm twittergasm when it was over, and it’s easy to see why” reading a post on Twitter. (www.chortle.co.uk). Taken from GloWbE. A feeling of enthusiasm produced when dealing with Vampiregasm something related to vampires, such as documentaries or films where they appear. “Looks like they’re scared that Cameron may pull A feeling of excitement another vetogasm surprise moment, especially with no Vetogasm produced by imposing a veto Lib Dem fifth columnist minister in the FO” on something or someone. (www.politicalbetting.com). Taken from GloWbE. Lexis, 14 | 2019 137 A feeling of pleasure or “The Italian accent is a vowelgasm that reflects the Vowelgasm excitement when listening to spectrum of Italic experience” (www.twmagazine.net). vowel sounds. Taken from EnTenTen15. “If Israel attacks Iran, John Bolton will have such a A feeling of excitement related Wargasm wargasm that his head might explode” to war. (UrbanDictionary). Presumably a feeling of excitement or enthusiasm “Upon Jeans reading those words, she may experience a Wordgasm when using or reading a word wordgasm” (UrbanDictionary). or when reading a text that somebody has written. Appendix 5 Table 5: Cause and effect blends modelled on ‘eargasm’ Word Definition Example “His sound is a majestic blend of overblown hyperbole Feeling of enthusiasm when and meaningless adjectives that actually have nothing to Bassgasm listening to the sound of a do with music. Each festy season he journeys across the bass. land inducing immersive cutting-edge bassgasms and killin’ it” (www.artoutside.org). Taken from EnTenTen15. Feeling of excitement when “North Korean Beliebers are going to have a Biebergasm listening to music by Biebergasm when they see what we’ve put together” Canadian singer Justin (www.ssy.org.uk). EnTenTen15. Bieber. A feeling of enthusiasm “Choirgasm : An Evening of Coral Pleasure” Choirgasm produced by listening to (www.oica.upd.edu.ph). Taken from EnTenTen15. the music of a choir. A feeling of excitement “2008 with Gaga in Miss Universe was your Gagasm when listening to music by gagasm” (www.gagadaily.com). American singer Lady Gaga. A feeling of excitement when listening to the sound “Dude i just bought jackson kelly and i think i also need a of a guitar. Occasionally, it Guitargasm change of pants because i just had a can be used to refer to the guitargasm” (UrbanDictionry). feeling of pleasure obtained while playing the guitar. “The Metalgasm comes about four minutes in, when A feeling of excitement Halford goes from death metal growls up through the Metalgasm when listening to metal octaves to a scream that would summon Thor himself” music. (www.metaltalk.net). Taken from EnTenTen15. Lexis, 14 | 2019 138 Feeling of excitement when listening to an opera piece “-Soprano/tenor *sings high C*. -Audience member: Operagasm or attending an opera *operagasms*” (UrbanDictionary). performance. A feeling of enthusiasm Popgasm when listening to pop music. “It doesn’t take any real prominence until the song A feeling of excitement bursts into a full out rockgasm after the arresting tremolo Rockgasm experienced when listening bass bridge with the unsettling line “One of these days to rock music. I’m going to cut you up into little pieces!”” (www.sourceaudio.net). Taken from EnTenTen15. A feeling of excitement “We are grateful that there is now a Soulgasm Music Soulgasm when listening to soul Label” (www.girlpr.motionforum.net). Taken from music. EnTenTen15. Appendix 6 Table 6: Experiencer blends ending in ‘-gasm’ Word Definition Example “The Carnivorgasm at The Vortex Bar & Grill (Atlanta) is one pound of grilled sirloin patties topped with Feeling of pleasure pulled smoked pork, bacon, sliced ham, turkey, Carnivorgasm experienced by a carnivore cheddar cheese and whiskey pimento cheese, covered (when they eat meat). in Vortex barbeque” (www.reviewtravel.info). Taken from EnTenTen15. “Meanwhile, the clown had another A feeling of pleasure Clowngasm clowngasm” (www.tenka.seiha.org). Taken from experienced by a clown. EnTenTen15. “’Wow man, where’d you find this restaurant?’ ‘I was Exploragasm, A feeling experienced by trying to go to Jenny’s house and had an explorgasm!’” Explorgasm an explorer. (UrbanDictionary) “We have the following two little tid bits of A feeling of excitement information, which is guaranteed to produce fangasms Fangasm experienced by a fan of from even the most casual ‘Star Wars’ fan” someone or something. (www.scifipulse.net). Taken from EnTenTen15. Femaleorgasm, A feeling of physical “FemaleOrgasm Sex Guide” (www.lending.estate). pleasure experienced by a Femgasm, Taken from EnTenTen15. Femmegasm female. Lexis, 14 | 2019 139 “Growing up with the Tintin books by Belgian artist A feeling of excitement or Hergé, hearing news of it being turned into a movie enthusiasm experienced by Geekgasm by Steven Spielberg and Peter Jackson gave me a a geek, a huge fan of geekgasm” (www.sugoistuff.net). Taken from someone or something. EnTenTen15. This is the name of a comic for adults. It probably Herogasm stands for the feeling of pleasure experienced by the hero. “Horgasm: This is going to be an in-depth look at the Feeling of pleasure or highs, the lows, the pressures and orgasmic releases Horgasm excitement experienced by that make up every damn frickin’ second of Torstein’s (Torstein) Horgmo. (Horgmo) existence” (www.whitelines.mpora.com). Taken from GloWbE. “you also have the power and the ability to ensure A feeling of pleasure you get the level of sexual satisfaction you want by Ladygasm experienced by a lady. using Ladygasm toys” (www.womenmasturbation.org). Taken from EnTenTen15. A feeling of pleasure Lesbigasm experienced by a lesbian. “The Mangasm Edge is sure to bring you to an A feeling of pleasure incredible orgasm, massaging both your prostate, and Mangasm experienced by a man. your balls with its intense vibrating power” (www.jerking-off.org). Taken from EnTenTen15. This word is used to refer to one of Shannon Moore’s signature moves while wrestling. The term stands Mooregasm for the feeling of excitement or pleasure experienced by (Shannon) Moore. A feeling of excitement or “As a HUGE Star Trek fan I am having a huge nerdgasm enthusiasm experienced by over today’s announcement Star Trek is coming back Nerdgasm a nerd, a huge fan of to TV!” (www.statestimes.net). Taken from something or someone. EnTenTen15. “As my Bitachon, my trust of Creator, grew, my need A feeling of excitement for control let go and the fuel of prayergasm has taken Prayergasm experienced by a prayer. me on a whirlwind tour of my devotional soul ever since!” (www.jewrotica.org). EnTenTen15. Lexis, 14 | 2019 140 “A prorgasm is what particularly keen programmers A feeling of pleasure or get when they have finally completed a program that Prorgasm excitement experienced by has been giving them grief for some time” a programmer. (www.meta.uncyclomedia.org). Taken from EnTenTen15. A feeling of pleasure experienced by a person Queergasm who identifies himself or herself sexually as being queer. “Walton tells of the Torygasm: Thatcher funeral will Feeling of excitement cost £10 million and involve 700 military personal. Big Torygasm experienced by Tories. Ben will be silenced” (www.cyberunions.org). Taken from EnTenTen15. A feeling of pleasure “Dude ,I just had sex with a Latino prostitute and she experienced by a woman Whoregasm totally had a Whoregasm all on my lap” who works in the (UrbanDictionary) prostitution business. Appendix 7 Table 7: Coordination blends ending in ‘-gasm’ Word Definition Example “I began using several of my favourite tantric techniques to A feeling of bliss circulate sexual energy between me and Sydney. Before I accompanied by a knew it, a little blissgasm shivered up my spine, followed by Blissgasm feeling of physical an actual clitoral orgasm... I was so amazed, I had to stop and pleasure. lean against a wall” (www.guardian.co.uk). Taken from GloWbE. A feeling of physical “This was very pleasurable, and triggered a series of deep pleasure occurring clitoral orgasms, accompanied by a burst of emotion, which I Crygasm simultaneously with call a crygasm” (www.sexecology.org). Taken from the activity of crying. EnTenTen15. This term is used to express something is “Have you seen this thing? This sexy macho bloated Hot someone’s dream and Wheels fantasia dreamgasm of a car-like drunken child’s Dreamgasm simultaneously funbot crayon sketch?” (www.commondreams.org). Taken produces a feeling from EnTenTen15. excitement in that person. Lexis, 14 | 2019 141 A feeling of joy one experiences Joygasm accompanied by a feeling of pleasure or excitement. A feeling of rage experienced by a “Rick was just having a little ragegasm at being caught lying. person, occurring Ragegasm There’s the shame of lying, coupled with the shame of together with a feeling dutifully doing so on behalf of the Party” (GloWbE). of pleasure or excitement. “when these people who take this drug yawn, they have an orgasm. They -- it’s called -- you know, I’m doing this as a A yawn which co- public service. I am -- I’m a highly trained broadcast Yawgasm, occurs with a feeling of specialist, ladies and gentlemen. This -- do not -- many of Yawngasm physical pleasure. you are thinking that I am advocating this kind of thing. I’m simply warning you. They’re calling this effect the yawgasm” (COCA). Appendix 8 Table 8: Adjectival blends ending in ‘-gasm’ Word Definition Example “Aurgasm seeks to bring you an eclectic This blend is formed by the Aurgasm menagerie of aural pleasures” (www.aurgasm.us). fusion of ‘aural’ and ‘orgasm’. (EnTenTen15). In this case, it is clear that the word orgasm keeps its original “The Fakegasm: where faking it until you make it Fakegasm meaning. It is fused with the doesn’t work. Ever fake an orgasm?” adjective fake to create a new (www.sexologyinternational.com). word meaning ‘a fake orgasm’. “Have you seen this thing? This sexy macho bloated Hot Wheels fantasia dreamgasm of a car- Faux + orgasm, meaning ‘a false/ Fauxgasm like drunken child’s funbot crayon sketch?” faux orgasm’. (www.commondreams.org). Taken from EnTenTen15. This expression is used to refer Forevergasm to an orgasm lasting for a very long period of time. Lexis, 14 | 2019 142 In this case, the adjective “the Surfing Goat Dairy where I had a gastronomical is used as modifier gastronogasm over the fresh flavored cheeses. Gastronogasm or orgasm. The word means ‘a (*Gastronogasm* I think I made up a new word!)” feeling of gastronomical (www.tobyneal.net). Taken from EnTenTen15. pleasure’. The source words of this blend Literarygasm are literary and orgasm. ABSTRACTS The aim of this work is to study the nature of blends’ splinters, elements that have been often referred to as ‘final combining forms’, whose status remains unclear. Our specific objective is to analyze blends containing the element ‘-gasm’ to check tendencies in their formation. We intend to prove that the sequence ‘-gasm’ possesses a high degree of productivity and is used recurrently in the formation of new words. Through the analysis of the bases to which the splinter is attached and the study of the relationship existing between the component parts of the blends, we aim at discovering which mechanism is at work, while attempting to accommodate the operation within the theoretical framework of Construction Morphology, an approach to morphology within the overarching theory of Construction Grammar. This approach, proposed by Geert Booij, acknowledges the existence of constructions as pairings of form and meaning at word level, and thus it considers that abstractions allowing speakers to coin new complex words are based on actual instances of words that are memorized. We have compiled our own 200-sample corpus of novel English blends coined by native speakers of English and taken from a variety of sources (online corpora, websites listing neologisms, social networks, including Twitter, and even online discussion threads). For the data analysis, we proceed as follows: (i) identify the cases of hapax legomena in the corpus in order to account for the productivity of the process; (ii) recover the source words of all the blends subject to study and contextualize the items in our corpus, by providing an instance of each word in a wider context; (iii) analyze the relationships existing between the first and the second component of the novel coinages, according to different factors, such as the notion of causality and the cognitive and experiential relationship of cause and effect, the syntactic behaviour of the source words and the semantic roles fulfilled by these units; (iv) identify blends exhibiting a similar behaviour and propose constructional schemas to account for their creation. The results of our analysis show that the splinter analyzed appears to be highly productive, and that blends ending in ‘-gasm’ can belong to different groups, the most significant of which are the following: (a) cause and effect blends, in which the relationship between the source words is based on the notion of causality. Two different subsets can be identified in this group, depending on the meaning that ‘-gasm’ acquires in the resulting blend: that of physical pleasure, as in toygasm or that of a feeling of excitement or enthusiasm, as in neologasm; (b) experiencer blends, such as clowngasm, where the first participant fulfils the semantic role of an experiencer, that is, the first participant experiences the feeling of pleasure; (c) coordination blends, that is, blended words whose source words stand in a relationship of coordination, such as crygasm; (d) adjectival blends, in which the first element premodifies the feeling denoted by the second element, as in fakegasm. Based on our results, constructional morphology, and especially constructional schemas, seem to be a very appropriate tool for explaining the formation of these lexemes. Le but de cet article est d’étudier la nature des fracto-morphèmes contenus dans les amalgames, ces éléments généralement connus sous le nom de ‘final combining forms’ en anglais, et dont le Lexis, 14 | 2019 143 statut demeure encore flou. Notre objectif principal est d’analyser les amalgames contenant l’élément ‘-gasm’ afin de mettre au jour des tendances quant à leur formation. Nous souhaitons montrer que la séquence ‘-gasm’ exhibe un haut degré de productivité et est fréquemment convoquée lors de la formation de nouveaux lexèmes. C’est à travers l’analyse des bases auxquelles est attaché le fracto-morphème et l’étude des relations existantes entre les parties constitutives des amalgames que nous souhaitons dégager le procédé à l’oeuvre, dans une optique théorique de morphologie constructionnelle, sous-branche de la grammaire de construction. Selon Geert Booij, cette approche postule l’existence de constructions en tant que couplage forme-sens au niveau du lexème ; par conséquent, cette approche part du principe selon lequel ce sont les abstractions qui permettent aux locuteurs de créer de nouveaux lexèmes complexes basés sur des exemples attestés de lexèmes qui ont été mémorisés. Pour construire notre corpus, nous avons ainsi recueilli un échantillon de 200 nouveaux amalgames inventés par des locuteurs natifs anglophones dans des contextes variés (corpus en ligne, sites internet répertoriant des néologismes, réseaux sociaux, dont Twitter, et également fils de discussion en ligne). Pour l’analyse des données, nous avons procédé de la manière suivante : (i) identification des cas d’hapax dans le corpus afin de verifier la productivité du procédé d’amalgamation ; (ii) mise au jour des termes sources de tous les amalgames étudiés et contextualisation de chaque amalgame dans le corpus en fournissant un contexte large pour chacun ; (iii) analyse des relations existantes entre le premier élément et le second élément des amalgames, selon divers critères tels que la notion de causalité et les relations cognitives et expérientielles de cause à effet, le comportement syntaxique des termes sources et les rôles sémantiques joués par ces unités ; (iv) identification des amalgames ayant des propriétés semblables et proposition d’un modèle constructionnel afin de rendre compte de leur création. Les résultats de notre étude indiquent que le fracto-morphème ‘-gasm’ exhibe une forte productivité et que les amalgames se terminant par ce même fracto-morphème appartiennent à divers groupes, dont les plus représentés sont les suivants : (a) les amalgames de type cause à effet pour lesquels la relation entre les termes sources est fondée sur la notion de causalité. Deux sous-groupes différents peuvent être identifiés, selon le sens que ‘-gasm’ revêt dans les amalgames : plaisir physique (toygasm) ou sentiment d’enthousiasme (neologasm) ; (b) les amalgames expérientiels (clowngasm) dans lesquels le premier actant joue un rôle sémantique expérientiel, c’est-à-dire qu’il éprouve un sentiment de plaisir ; (c) les amalgames coordinatifs, c’est-à-dire ceux dans lesquels les termes sources sont dans une relation de coordination (crygasm) ; (d) les amalgames adjectivaux dans lesquels le premier élément prémodifie le sentiment évoqué par le second élément (fakegasm). D’après nos résultats, la morphologie constructionnelle et, plus particulièrement, les modèles constructionnels se révèlent être très pertinents pour mettre au jour la formation de ces lexèmes. INDEX Keywords: lexical blending, blend splinters, constructional morphology, constructional schemas, semantics Mots-clés: amalgamation lexicale, fracto-morphèmes, morphologie constructionelle, modèles constructionnels, sémantique Lexis, 14 | 2019 144 AUTHOR ALEJANDRO BARRENA JURADO Independent researcher [email protected] Lexis, 14 | 2019 145 Headedness in contemporary English slang blends Gorica Tomić The paper is a part of the research done within the project no. 178014 The Dynamics of the Structures of the Contemporary Serbian Language, funded by the Ministry of Education, Science and Technological Development of the Republic of Serbia. Introduction The importance of investigating minor word-formation processes (e.g. acronyms, blends, clippings, etc.) in various registers, of which slang with its great potential for morphological, lexical, and semantic innovation seems to be constantly shaping the standard language, as well as a number of non-standard ‘lects’ [Mattiello 2005: 7], has been emphasized more than once in the relevant literature [Mattiello 2008: 43, 59; Müller et al. 2015: 213].1 Furthermore, slang and informal register in general have often been considered a rich source of blends (cf. Pound [1914: 2]; Adams [1973: 157]; Ralli and Xydopoulos [2012: 47–48]; Ronneberger-Sibold [2012: 124]; Mattiello [2008: 65–66]; Matiello [2013: 232]. Therefore, the primary motivation for the present research is found in the works by Müller et al. [2015: 213], where the importance of systematically exploring different registers for their specificities in the context of word-formation is stressed, and Mattiello [2008], in which the focus of attention is, inter alia, on a description of slang morphology and its peculiarity. The theoretical framework adopted in this paper for the analysis of a number of contemporary English slang blends is the one of Extra-grammatical Morphology (EM). According to Dressler [2000: 1], EM “is the antonym of morphological grammar”, in that various extra-grammatical operations it includes (e.g. abbreviation, blending, clipping, etc.) “are not clearly identifiable and their input does not allow a prediction of a regular output” [Mattiello 2013: 1].2 That is, From a morphological point of view, these processes are considered unpredictable, in the sense that we cannot predetermine how much of the original lexeme will be retained in the new formation […]. [Mattiello 2013: 4] Lexis, 14 | 2019 146 Within EM a blend can therefore be defined as a deliberate creation of a new word out of two (or rarely more) previously existing ones in a way which differs from the rules or patterns of regular compounding. [Ronneberger-Sibold 2006: 157; cf. also Dressler 2000: 5] Blends are also referred to as “unpredictable formations”, i.e. the products of a creative technique which deviates from productive rules of grammatical morphology and are thus less natural than, for instance, compounds [Marchand 1969: 451; Bauer 1983: 232; Dressler 2005: 268; Mattiello 2008: 16, 25, 138; Gries 2012: 145; Körtvélyessy 2014: 296]. Although blends represent “non-rule governed morphological innovation”, there are, however, certain regularities or preferences in their formation, as well as similarities between them and some grammatical morphological processes, such as compounds [Marchand 1969: 451; Thornton 1993: 147; Plag 2003: 122, 125; Ronneberger-Sibold 2006: 155; Lehrer 2007: 116; Mattiello 2008: 138–139; 2013: 5–6, 14, 131; Gries 2012: 146; Bauer et al. 2013: 458, 462; Olsen 2014: 46; Renner 2015: 123; 2019: 36]. Furthermore, Ronneberger-Sibold [2010: 206] claims that such “creative techniques” are highly motivated by the reduction of the transparency of the output while retaining an optimal form for it. She [Ronneberger-Sibold 2010: 206] also states that there are “certain […] communicative contexts which favor totally or partially opaque words labelling their referents, rather than transparent ones describing them […]”. Such contexts include, inter alia, “all secret languages” [Ronneberger-Sibold 2010: 206], among which, as we will see shortly, slang holds an important position. Similarly to blends, slang is more often than not considered of minor importance, compared to standard language, “created and […] used by those beyond the social, and by extension linguistic, pale” [Green 2016: 9]. Slang has also been rather incorrectly conceptualized by many lexicographers as necessarily “informal or bad language” [Mattiello 2005: 7, 10–11; 2008: 31, 32]. However, there are authors (cf. Mattiello [2005]; [2008]) who have quite successfully managed to capture and demonstrate not only the pervasiveness of the phenomenon of slang across speech, but also its (creative) value, i.e. the originality of its forms and meanings (the quality I also endeavor to explore and demonstrate in this paper). According to Mattiello [2005: 9], one of the two main reasons (the other being its ephemerality) why the concept of slang is almost impossible to define, at least not properly, is its “rather wide, all-encompassing nature”. Namely, it has been this very nature that has led many sociolinguists to consider slang equivalent to such non-standard language varieties as cant, jargon, dialect, vernacular, or even accent [Mattiello 2005: 7, 9, 11]. With regard to the nature of slang, Green [2016: 24] believes that one of the most prominent characteristics of slang is its subversive nature. That is, the greater part of its lexis is based on the recycling of terms that are well established in standard use. Slang takes them over, turns and twists them, and offers them up in new combinations and senses. [Green 2016: 27; cf. also Mattiello 2005: 8, 12; 2008: 59, 156] Another important characteristic of slang is its secrecy [Bareš 1974: 183; Mattiello 2005: 13, 27]. Even though the etymology of the word slang as a blend of secret and language has been abandoned [Green 2016: 29], one cannot de-emphasize “conscious secrecy in the formation and use of slang” or its resulting semantic indeterminacy [Mattiello 2005: 17, 23, 25; 2008: 16–17]. For instance, referring to the semantics of slang in general, Mattiello [2008: 44–45] reports that there is a tendency in slang to “name things indirectly or figuratively, especially through metaphor […], metonymy […], synecdoche […], euphemism […], and irony […]” (cf. also Sornig [1981: 61]). Lexis, 14 | 2019 147 Furthermore, slang words are, similarly to blend words, often characterized as ephemeral and/or trendy [Lehrer 2003: 369; Mattiello 2005: 9; 2008: 47; 2013: 8; Green 2016: 30, 85]. Considering all the above, and especially the fact that both blends and slang are still controversial issues, the aim of the present paper is to investigate some of the formal and semantic characteristics of blends in contemporary English slang, or more specifically the patterns by which they are formed, their morphosyntactic (i.e. grammatical) and semantic headedness, as well as the relationship between the two types of head. Accordingly, the paper is structured as follows: Section 1 gives a brief account of some of the formal and semantic characteristics of English blends in general and slang blends in particular. It also provides some definitions and considerations of the notions of ‘head’ and ‘headedness’ in morphology. Section 2 explains the data collection and methodology. Section 3, which is divided into three subsections, presents a thorough qualitative and, to a lesser degree, quantitative analysis of the patterns employed in the formation of the slang blends, as well as the analysis of their morphosyntactic and semantic headedness. Conclusion offers some general remarks on the analyzed slang blends, including the relationship between the two types of head. 1. On some formal and semantic characteristics of blends in English According to Pinker [2007: 297], by far the most common source of new words is the use of existing ones, either in their full or clipped form. In producing these new words, each language has its own more or less predictable mechanisms. One such, but rather unpredictable mechanism which has become particularly fashionable in English over the last few decades is most often referred to by linguists as blending. The lexical phenomenon of blending is said to have been popularized by Lewis Carroll in his nonsense poem Jabberwocky (1871) [Pound 1914: 1]. Ever since then, many attempts to define blending and its products – blends have been made. For instance, under one of the well-known and oft-cited definitions, a blend is defined as “a new lexeme formed from parts of two (or possibly more) other words in such a way that there is no transparent analysis into morphs” [Bauer 1983: 234]. On the basis of one other definition, blends are complex words whose formation “involves two or (rarely) more base words” in such a way that these base words lose some of their phonetic (or orthographic) material and are thus “best described in terms of prosodic categories” [Plag 2003: 121]. Similarly, Bauer [2003: 225] claims that in a blend the first part of one word is combined with the last part of another, in such a way that it is usually no longer (in terms of number of syllables) than the longer of the two base words. What is more, the switch point between the two words seems to be determined by the phonology of the bases, so that, in slang + language > slanguage, we avoid the phonological repetition of /læŋ/. The influence of prosody on the formation of blends is also mentioned by Olsen [2014: 46] who states that in blending, two lexemes are combined, but at the same time they are superimposed upon one another leading to a shortening of one or both constituents. Nevertheless, the meaning of each constituent lexeme flows into the Lexis, 14 | 2019 148 meaning of the blend in the same manner as with compounds […]. The shortened forms of the blend’s constituents are subject to prosodic factors.3 A somewhat more elaborate definition of blends is provided by Mattiello [2013: 6]. She defines blends as words obtained by fusing parts of at least two source words, at least one of which is curtailed and/or there is a graphemic/phonemic overlap between them. Hence, they exhibit some sort of structural fusion, which is reflected in their semantics […]. Finally, “the principles of maximization and economy of effort”, which seem to be crucial for the creation of blends too, are emphasized by Balteiro [2013: 886]. In other words, blends “allow the creator to express in a single word what otherwise would formally take at least two words” [Balteiro 2013: 886]. Blends also meet “the brain’s need for a denser information load by shortening certain very familiar concepts which require a shorter processing time” [Chung 2009, as cited in Balteiro 2013: 886]. Regarding slang blends in particular, Mattiello [2005: 22] claims that they “may exhibit some anomalies”, in that certain slang blends are formed by some non-prototypical patterns. Therefore, she [Mattiello 2005: 22; 2008: 139–141] distinguishes prototypical from partial (and less typical) blends. In prototypical slang blends, the first part of one word and the end of another are fused, whereas in partial slang blends, one of the input words is retained in its entirety, or one word is inserted within another one, or the two words simply overlap, as in kidvid, gaydar, gazunder, dirty-mac, ambisextrous, nerk, or boolivan. Some slang blends represent a mix of an acronym or an initialism and (part of) a regular word, such as Amerikkka or buppie. Also, most of prototypical slang blends belong to the syntactic class of adjectives and are formed from other (non-)standard adjectives which are similar or related in meaning. The pattern is repeated in some prototypical slang nominal blends as well. Besides the above-mentioned semantic similarity, there is another significant characteristic on the basis of which blends are comparable to compounds. It is the notion of a ‘head’ [Kubozono 1990: 1]. Namely, the notion of a ‘head’ is used in morphology to distinguish between endo- and exocentric compounds [Halupka-Rešetar and Lalić-Krstin 2009: 119]. Endocentric compounds are considered hyponyms of one of their members, whereas exocentric ones are not hyponyms of either of their members (e.g. names of people, animals, and plants) [Bauer 2003: 42, 177; Plag 2003: 146].4 According to Williams’ Righthand Head Rule (RHR), “the head of a morphologically complex word” in English is “the righthand member of that word” [Williams 1981: 248]. 5 This claim, however, has proved to be rather problematic, especially in the context of compounds, given the number of exceptions which include not only exocentric compounds, but also those that have their head on the left or those whose members equally contribute to the resultant meaning and can thus be said to be double-headed [Marchand 1969: 13–14; Bauer 2003: 179, 182; 2008; Plag 2003: 147, 183; Bauer et al. 2013: 28, 635]. It is also noteworthy that there are numerous exceptions to endocentric compounds in slang and that the notion of ‘headedness’ is a controversial issue in blends too, since the head may be either the right-hand or the left-hand member, or they may even be termed “headless” [Mattiello 2008: 85; 2013: 60, 129–130]. Therefore, it seems more appropriate to say that the head of an endocentric compound (and by extension an endocentric blend) is the member which transfers its grammatical or semantic information to the compound or blend as a whole [Bauer 2003: 177]. Lexis, 14 | 2019 149 With regard to the semantic headedness of blends, Bat-El [2006: 66] claims that “a blend is one word that delivers the concept of its two base words and its meaning is thus contingent on the semantic relation between the two base words”. She [Bat-El 2006: 66] illustrates the aforesaid by the blend skinoe which is made up of ski and canoe, where the latter “functions as the semantic head, since skinoe is a type of canoe”. But, in snazzy, “neither snazzy nor jazzy functions as a head and the meaning of the blend is thus a hybrid of the meaning of the two (sometimes near-synonymous) base words”. Bat-El [2006: 67] further remarks that blends in English “do not show preference for endo- or exocentric relation, whereas compounds are mostly endocentric”, but that endocentric blends, similarly to endocentric compounds, are largely right-headed (cf. also Gries [2012: 164]). In exocentric blends, however, the base words are of equal semantic status [Bat-El 2006: 67].6 Finally, Bat-El [2006: 67] mentions that for some blends such as smog or brunch it is not quite clear which semantic relation exists between its base words.7 In their analysis of the semantics of blends, Bauer et al. [2013: 485] also mention a small group of blends “that is difficult to characterize”, in that some of those blends are rather opaque (e.g. Boyzilian, Internot, transwestite) and thus require extra semantic information to be properly interpreted. The authors [Bauer et al. 2013: 483–484] also claim that blends are semantically remarkably similar to non-argumental attributive and coordinative compounds, that is, in semantically attributive blends, as in attributive compounds, the blend as a whole is a hyponym of the second blended element, and the first element bears some contextually plausible relationship to the second. […] For blends that have coordinative interpretations, we find both appositive and compromise types. […] As with compounds of this sort, the appositives denote the intersection of two types of entity or action. […] The compromise coordinatives denote hybrid entities or concepts. Another extremely useful typology of blends is proposed by Ronneberger-Sibold [2006: 168–169]. This typology, which is based on the analysis of the transparency of German blends in satirical texts and brand names, includes: complete blends, contour blends, semi-complete blends, and fragment blends (given from most to least transparent). Complete and contour blends involve two subtypes each. I will return to this typology in Subsection 3.1. and Conclusion, since one of the aims of the present paper is to investigate the slang blends’ morphotactics. 2. Data collection and methodology The examples of contemporary English slang blends, in which two input words are combined in such a way that at least one of them is shortened and/or there is a phonemic and/or graphemic overlap between them or their parts [Mattiello 2013: 6], are collected from the online version of Green’s Dictionary of Slang (GDoS).8 The primary reason why I decided to make use of a dictionary in collecting the data for the present research is the fact that most dictionaries (along with definitions) readily provide valuable information about the date of first use of a word, its origin, examples of use, etc. The reason why I decided to use an online version of a dictionary of slang instead of a printed one is the fact that it is updated more regularly than the printed version, thus providing its users with an invaluable insight into contemporary language use, especially if a language is as dynamic as English is and all the more so if the research is aimed at a chameleon-like register like slang. Lexis, 14 | 2019 150 I chose to use this specific online dictionary of slang among many other similar dictionaries available primarily because, at least to my knowledge, it seems not to have been exploited extensively in slang research thus far and also because it offers some exceptionally advanced search tools.9 These, for instance, include search not only by word history, meaning, and usage, but also by the time period the word is actively used in. Once the search term is selected, users can further refine the search by choosing one of the following: period of use, date of first use (word), or date of first use (sense). Since I am interested in analyzing certain slang blends only, i.e. those whose date of first use is recorded between 2000 and 2019, this tool proved to be a great help. Actually, the original idea was to include only the blends from the 21st century, but their number was unsatisfactorily low that I had to extend the search period for an additional year, i.e. 2000. This way, I managed to create a data set consisting of 60 blends (61 senses). It is noteworthy that within this time period (2000–2019) the most prolific years, in terms of the number of the blends extracted, are 2000, 2001, and 2003. The actual use of these slang blends is authenticated by fully-referenced citations in GDoS. In those rare cases where there were no such attestations available, I searched other online dictionaries of the English language such as the Urban Dictionary (UD), the Collins English Dictionary (CED), the Merriam-Webster Dictionary (MWD), and the Oxford English Dictionary (OED), as well as Google for examples of their use in authentic text stretches. 3. Data analysis and discussion 3.1. Formal characteristics of the slang blends The collected slang blends are first analyzed as regards the patterns employed in their formation. Namely, 9 formation patterns (a)–(i), given below from most to least frequent, are identified. The blends within each formation pattern are given in alphabetical order. The (parts of the) input words the blends are made up of are italicized and annotated for the syntactic category they belong to. If there is an overlap involved in the blend, the overlapping segments are indicated by being underlined.10 The meanings of the slang blends, as well as some examples of their contextual usage, will be provided in Subsections 3.2. and 3.3., in which I intend to investigate blends’ morphosyntactic and semantic headedness in detail. All examples of the blends’ use in context, as well as their meanings are from GDoS, unless otherwise specified. (a) The first part of the first word is blended with the second part of the second word, with a possible (often minimal) phonemic and/or graphemic overlap between the fragments – 26 blends. Blends which manifest a phonemic overlap are (03), (12), (17), and (23); those which manifest a graphemic overlap are (06), (15), and (16); and those which manifest a phonemic-cum-graphemic overlap are (01), (04), and (11). In Ronneberger-Sibold’s typology [2006: 169], this type of blend where only fragments of input words are employed in its formation is appropriately termed fragment blends. The term also applies to the patterns in (d) (only the first blend and the third one, though), (f), and (i). It is the most opaque type in her classification. Examples are: (1) basticle11 n. [bastard n. + SE12 testicle n.] (2) blamps n. [SE big adj. + headlamps n.] (3) bonerific13 adj. [bonaroo adj. + SE terrific adj.] Lexis, 14 | 2019 151 (4) chunt n. [chump n. + cunt n.] (5) Crentley14 n. [Chrysler n. + Bentley n.] (6) darty n. [daytime n. + party n.] (7) Datto n. [Datsun n. + auto n.] (8) datty adj. [dotty adj. + batty adj.] (9) dopalicious15 adj. [dope n. + -a- + SE delicious adj.] (10) faburrific adj. [SE fabulous adj. + terrific adj.] (11) fratastic adj. [fraternity n. + fantastic adj.] (12) hunty n. [honey n. + cunty n.] (13) knack16 adj. [naff adj. + wack adj.] (14) Koreegro n. [SE Korea n. + negro n.] (15) neek n. [nerd n. + geek n.] (16) nugger n. [nugget n. + nigger n.] (17) pootenanny n. [punaany n. + hootenanny n.] (18) shart v. [shit v. + fart v.] (19) slore17 n. [SE slut n. + whore n.] (20) spleefer n. [spliff n. + reefer n.] (21) spum v. [SE sperm n. + cum v.] (22) spuzz18 n. [SE sperm n. + jizz n.] (23) stray19 n. [straight adj. + gay n.] (24) wanksta20 n. [wannabe adj. + gangsta n.] (25) wegro n. [SE white n. + negro n.] (26) widdle n. [wee n. + piddle n.] (b) The entire first word is blended with the second part of the second word, with a possible (often minimal) phonemic (and graphemic) overlap – 14 blends. A phonemic overlap is attested in blends (27), (30), (35), and (40), while a phonemic-cum-graphemic overlap is attested in those in (29), (34), and (37). In Ronneberger-Sibold’s typology [2006: 169], this type, as well as the ones in (c), (g), and (h) are appropriately termed semi-complete blends, since they contain one full word in the output. These are, therefore, less opaque than the ones in (a), (d), (f), and (i), containing no full words. Examples are: (27) Blaxican21 n. [SE B/black n. + Mexican n.] (28) budiquette n. [bud n. + SE etiquette n.] (29) cashish n. [SE cash n. + hashish n.] (30) chillax v. [chill (out) v. + SE relax v.] (31) craptageous adj. [crap adj. + advantageous?22 adj.] (32) craptard n. [crap n. + retard n.] (33) craptastic adj. [crap adj. + fantastic adj.] (34) friendscape23 v. [friend n. + landscape v.] (35) Jailic24 n. [SE jail n. + Gaelic n.] (36) jerkitude n. [jerk n. + attitude n.] (37) kidult n. [kid n. + SE adult n.] (38) mackadelic25 adj. [mack n. + -a- + psychedelic] (39) mangina n. [SE man n. + vagina n.] (40) twoonie n. [SE two number + loonie n.] (c) The first part of the first word is blended with the entire second word, with a possible (phonemic and) graphemic overlap – 9 blends. The overlapping blends include cases which manifest a graphemic overlap, as in (42) and (49) or a phonemic-cum- graphemic overlap, as in (41) and (45). Examples are: (41) buffugly26 adj. [butt n. + fugly adj.] Lexis, 14 | 2019 152 (42) cramazing adj. [crazy adj. + SE amazing adj.] (43) gertoss27 n. [SE girl n. + toss (off) v.] (44) grape n. [gang n. + rape n.] (45) mexicoon n. [SE Mexican n. + coon n.] (46) schmiddy n. [SAusE28 schooner n. + middy n.] (47) shwasted adj. [shitfaced adj. + wasted adj.] (48) sororowhore n. [SE sorority n. + -o- + whore n.] (49) twasted adj. [twatted adj. + wasted adj.] (d) (The initial or final part of) the second word is inserted within the part of the first one, with the possibility of a phonemic-cum-graphemic overlap, as in (51) and (53) or a discontinuous graphemic overlap, as in (52) – 4 blends. Examples are: (50) cadazy adj. [SE crazy adj. + mad adj.] (51) hangry adj. [hungry adj. + angry adj.] (52) scrav v. [SE scavenge v. + scrape up v.] (53) yestergay n. [SE yesterday n. + gay n.] (e) Both input words survive due to an obligatory phonemic overlap (54) or an obligatory phonemic-cum-graphemic overlap, as in (55) and (56) – 3 blends. This type of blend is referred to by Ronneberger-Sibold [2006: 168] as complete blends, or more specifically, its subtype telescope blends. It is the most transparent one in her typology. As evidenced by only three examples below, this type is one of the least frequent in my data set. This may be due to the slang register the blends originated in, i.e. slang’s preference for indeterminacy. (54) Blinglish n. [bling n. + SE English n.] (55) bromance n. [bro n. + SE romance n.] (56) requestion n. [request n. + question n.] (f) The first part of the first word is blended with the first part of the second word – 1 blend: (57) smim29 n. [spastic n. + mimic n.] (g) The entire first word is blended with a medial part of the second word – 1 blend: (58) shegarry n. [sheg v. + carry-on n.] (h) The first part of the second word is inserted within the entire first word, with a minimal phonemic-cum-graphemic overlap – 1 blend: (59) slock v. [SE sock n. + slug v.] (i) The second part of the first word is blended with the second part of the second word – 1 blend: (60) Vaalie n. [Transvaal n. + japie n.] 3.2. Morphosyntactic headedness of the slang blends The analysis of morphosyntactic headedness of 60 slang blends, for the purpose of which I also consulted The Cambridge Grammar of the English Language [Huddleston and Pullum 2002], as well as some of the online dictionaries mentioned in the previous Lexis, 14 | 2019 153 section, shows that over half of them (i.e. 33) inherit their grammatical properties from the right-hand member, while only 6 blends, namely basticle, cadazy, gertoss, hunty, pootenanny, and shart have these properties determined by the left-hand member. There are also 21 blends that may be considered morphosyntactically double-headed owing to the fact that they inherit grammatical features from both of their input words. Consider first the morphosyntactically left-headed blends, such as the common, countable, and animate noun basticle30. Its input words (bastard and testicle) are also common and countable nouns. They, however, differ from each other, in that the former belongs to the subclass of animate nouns (the same way that the blend does), whereas the latter is part of the subclass of inanimate ones.31 Another blend that lends itself to a convincing analysis as morphosyntactically left-headed is the noun gertoss, for it is the left-hand member (girl) that is also a noun; the right-hand one being a verb (toss (off)). Unlike basticle and gertoss, the nominal blend hunty, used as a term of (sarcastic) endearment32, seems to evade such a straightforward left-headed interpretation, since it is made up of two other countable nouns, an endearment honey and cunty (‘(offensive) a woman considered sexually’ (CED)). However, if their distribution in a sentence is considered, the former normally realizes a vocative function [Huddleston and Pullum 2002: 522], the same way that the blend as a whole does, while the latter, based on the examples of its use provided in GDoS, does not.33 Slightly less problematic to analyze is the nominal blend pootenanny, used to refer to the female genitals. Although both of its input words are nouns, only the first one (punaany) is inanimate, whereas the second one (hootenanny) is utilized as ‘a general term of abuse’. Regarding the adjective cadazy, it involves blending of the two other adjectives (crazy and mad). However, if this adjectival blend is inflected for degree, it behaves the same way that crazy does, since its comparative and superlative forms require a formal change from y into i before being marked by the inflections -er and -est, respectively. This little nuance seems to allow us to categorize the resulting blend as grammatically left-headed. Finally, the verb shart is considered grammatically left-headed because, like its left-hand member (shit), it can be used not only intransitively (e.g. ‘He sharted at the party last night’ (Google) ), but also as a reflexive verb (e.g. ‘Wyatt sharted himself’(Google)). The slang blends whose both members can be taken to more or less equally contribute to the grammatical properties of the output, i.e. morphosyntactically double-headed blends, include: cashish, chillax, chunt, Crentley, datty, hangry, kidult, knack, neek, nugger, requestion, scrav, shwasted, slore, smim, spleefer, spuzz, twasted, Vaalie, wegro, and widdle. For instance, the blend Crentley, which belongs to the class of countable and inanimate nouns, is taken to be grammatically double-headed because both of its input words, namely Chrysler and Bentley are also countable and inanimate nouns. The blends cashish, spuzz, and widdle are, like their members, uncountable and inanimate nouns. Similarly, spleefer is an inanimate, but countable noun whose both input words (spliff and reefer) are also inanimate and countable nouns. The countable, but abstract nominal blend requestion can be said to inherit these two characteristics from either of its input words, since both request and question are abstract and countable nouns. Another, highly representative example of a morphosyntactically double-headed blend is kidult. Namely, the blend as a whole is, like its members (kid and adult), a common, countable, Lexis, 14 | 2019 154 and animate noun. The same applies to the following nominal blends: chunt, neek, nugger, slore, smim, Vaalie, and wegro. With regard to the adjectives datty, hangry, knack, shwasted, and twasted, they are as gradable as the adjectives they are made up of, namely dotty and batty, hungry and angry, naff and wack, shitfaced and wasted, twatted and wasted, respectively. Finally, the intransitive blended verb chillax is interpreted as grammatically double-headed, since its input words are also intransitive verbs, namely chill (out) and relax.34 Another verbal blend that is taken to be grammatically double-headed here is scrav ‘(UK juv.) to borrow or steal (usu. money)’, since it is made up of the other two transitive verbs scavenge and scrape up. As already pointed out, the majority of the blends in my data set are morphosyntactically right-headed. Such examples include: blamps, Blaxican, Blinglish, bonerific, bromance, budiquette, buffugly, cramazing, craptard, craptastic, craptageous, darty, Datto, dopalicious, faburrific, fratastic, friendscape, grape, Jailic, jerkitude, Koreegro, mackadelic, mangina, mexicoon, schmiddy, shegarry, slock, sororowhore, spum, stray, twoonie, wanksta, and yestergay. Now, consider first the nominal blend bromance. Both of its input words are also nouns, but unlike the noun on the left (bro), the one on the right (romance) is abstract the same way that the blend as a whole is. The abstract nominal blends Blinglish, budiquette, Jailic, grape, and jerkitude can all be interpreted in more or less the same manner, since their right-hand input words, namely English, etiquette, Gaelic, rape, and attitude are all abstract nouns, too. Their left-hand members, however, belong to various subclasses, notably uncountable nouns (bud, jewelry), common and countable nouns (jail), or countable and animate nouns (gang, jerk). The other nominal slang blends, such as craptard, Koreegro, sororowhore, and yestergay are all animate, the same way that their right-hand input words retard, negro, whore, and gay are. The opposite is, however, true of the noun mangina (‘(gay) anus’), whose right-hand member is inanimate the same way the blend as a whole is, whereas the left-hand one is animate. Slightly more problematic to characterize from the perspective of morphosyntactic headedness is the nominal blend Datto. Although its left-hand member (Datsun) is a proper noun (i.e. the Datsun Motor Company), the common right- hand one (automobile) appears to be grammatically more dominant, since the whole blend is simply a common noun denoting a kind of automobile (e.g. “Then I reckon you left the Datto at the station, caught the first train to the city.”) [cf. Huddleston and Pullum 2002: 522]. On the other hand, the nominal blend shegarry lends itself to a straightforward analysis as morphosyntactically right-headed, since only its right-hand member is a noun (carry-on); the left-hand one being a verb (sheg). Similarly, the verb friendscape inherits its grammatical properties from the right-hand member (i.e. the verb landscape), the left-hand one being a noun (friend). Examples of right-headed blends in which the two input words also differ from each other categorially are: blamps, buffugly, dopalicious, fratastic, slock, spum, stray, twoonie, and wanksta. The nominal slang blend mexicoon is, like its right-hand member (coon), a common noun; the left-hand one (Mexican) being a proper noun. In the noun Blaxican, both input words – B/black and Mexican – are animate and countable nouns, as is the blend. However, unlike its left-hand member, which can be used either as a proper or a common noun (cf. OED), the blend as a whole is part of the subclass of common nouns, as is its right-hand member. A useful factor in determining the grammatical head of the blend can be its plural form, more precisely the suffix it is marked by when pluralized. Lexis, 14 | 2019 155 For instance, when inflected for plural, the noun schmiddy behaves like its right-hand input word middy, in that there is an obligatory formal change from y into i when -es is to be added. Finally, the adjective cramazing is categorized as morphosyntactically right-headed, since its comparative and superlative forms are periphrastic or analytic, i.e. marked by more and most, as is the case with amazing, but not with crazy. An interesting case is the blended adjective bonerific, since both of its input words (bonaroo and terrific) are also adjectives.35 What makes the morphosyntactic head of this blend all the more difficult to ascertain is the fact that bonerific is gradable and is used in either of two functions, attributive and predicative, the same way both of its inputs are. However, when an adverb is to be formed, bonerific behaves like its right-hand member terrific, in that it takes -ally (not -ly). This seems to render the whole blend closer to terrific and thus grammatically right-headed.36 This also seems to be the case with the adjective – faburrific (*faburrificly). 3.3. Semantic headedness of the slang blends Despite being equally (if not more) demanding to determine, semantic headedness of the slang blends seems to provide far more illuminating insights into their rather complex nature. On the basis of the semantic analysis of 61 senses, all the slang blends in my data set are divided into endocentric (55) and exocentric (6). The endocentricity and exocentricity of the slang blends are established by applying a less restrictive hyponymy test [Bauer 2008; Mattiello 2008: 160, 166; Renner 2019: 36–37]. Among the endocentrics, 32 are single-headed (21 right- and 11 left-headed) and 23 are double- headed or semantically coordinate. Apparently, the exocentric blends are in the sizeable minority, which is to be expected when a less restrictive test of hyponymy is applied. In what follows, I will first analyze and discuss the dominant, endocentric group of the slang blends and then a small group of exocentric ones. Also, the following analysis of semantic headedness of the slang blends is primarily qualitative, since my data counts only 61 senses. However, some quantitative analysis is still done and presented here for the sake of completeness, but also as an attempt to learn if there are any tendencies regarding semantic headedness of blends in this specific register. Within the endocentrics, a further distinction can be made between those with the semantic head on the right (21 blends) and those with the semantic head on the left (11 blends). Semantically right-headed examples include: blamps, Blinglish, budiquette, buffugly, darty, Datto, dopalicious, fratastic, friendscape, grape, Jailic, jerkitude, kidult, mackadelic, mexicoon, shegarry, slock, sororowhore, spum, twoonie, and wanksta. Consider first the nominal blend Blinglish.37 Namely, based on its meaning ‘ (UK black/teen) Jamaican patois as adopted by white English youth’38, Blinglish is a form of spoken English and not of bling(-bling), which is used to denote ‘expensive, ostentatious clothing and jewellery’ (OED) . The noun bling(-bling), therefore, acts as a kind of modifier here. Other examples such as: budiquette (‘(US drugs) the etiquette that governs the smoking of marijuana’)39, buffugly (‘(US black) extremely unattractive’), darty (‘(US campus) a daytime party’)40, Datto (‘(Aus.) a Datsun automobile’), dopalicious (‘(US black) wonderful’)41, fratastic (‘(US campus) pertaining to fashions as espoused by a member of a fraternity’)42, grape (‘(N.Z.) gang rape’), Jailic (‘(Irish) Irish as learned while imprisoned in the prison at Long Kesh, Belfast’)43, jerkitude (‘social ineptitude, gaucherie’), kidult (‘an adult person who indulges in entertainment, usu. film, videotape or television, geared to attract both child and adult audiences’), mackadelic (‘self-promoting, parading Lexis, 14 | 2019 156 the qualities of a pimp’), slock (‘(US prison) to hit someone with a heavy padlock concealed in a sock’)44, twoonie (‘(Can.) a two-dollar coin’), and wanksta (‘a person who acts or dresses like a gangster but who is not involved in crime’ (CED)) can be interpreted the same way, since their left-hand members closely modify the (non-)figurative meaning(s) of their right-hand members or the heads. In a similar way, the noun shegarry (‘(UK black) annoyance, irritation’)45 and a verbal blend spum (‘(US black) to ejaculate’) are interpreted as semantically right-headed, owing to the fact that they are synonymous with their right-hand input words, carry-on and cum, respectively. The two blends, however, slightly differ regarding the relationship that exists between their heads and non-heads. Specifically, in shegarry, the non-head (sheg ‘to annoy, to provoke’) seems to amplify the meaning of the head, whereas in spum, the non-head receives an object interpretation. The noun blamps (‘large breasts’), on the other hand, does not lend itself to such an undemanding endocentric analysis, since the interpreter has to be familiar with the figurative, i.e. metaphoric slang meaning of the right-hand input word headlamps – ‘the female breasts’. Only in this way is it possible to consider the blend under discussion semantically endocentric in nature. Similarly, mexicoon ‘(US black) a black man who pursues Hispanic women’ is taken here to be a hyponym of its right-hand member coon which is used figuratively to mean ‘a black person’. Yet another blend that appears somewhat more difficult to characterize as semantically right-headed is the noun sororowhore, glossed as ‘(US campus) a derog. term for a female student who enjoys an active social life’. Based on its meaning, as well as the meaning of the noun whore – ‘derogatory a prostitute; A woman who has many casual sexual encounters or relationships’ (OED), the blend cannot unambiguously be interpreted as semantically right-headed, since an active social life does not necessarily involve ‘having or characterized by many transient sexual relationships’ (OED). But considering the semantic feature ‘promiscuous, or consisting of a wide range of different things’ (OED) the noun whore transfers to the output, sororowhore may be taken to be the semantic head of the whole blend. Similarly, if one applies the standard hyponymy test to the blended verb friendscape meaning ‘(US) to amend one’s circle of friends as listed on social media’46, the blend can’t be said to be semantically right-headed, since to friendscape is not a hyponym of to landscape. However, considering the above meaning of the blend, the right-hand member (i.e. the verb to landscape) has obviously transferred one of its semantic features, namely the seme ‘modify’ (MWD) to the output, thus rendering the blend’s right-headed analysis possible. All in all, the previous few examples of slang blends support the claims made in the literature about “the obscure relationship between some lexemes and their slang meaning(s)” which activates “the complicated cognitive processes” on the part of the interpreter [Mattiello 2008: 59]. More or less the same is true of the following semantically left-headed slang blends. Examples include: basticle, cashish, craptastic, craptageous , Crentley, gertoss, hunty, pootenanny, scrav, stray, and wegro. Take, for instance, the nominal blend basticle which, according to the Urban Dictionary (UD), is used to refer to ‘a jerk’. Considering the meaning of the blend’s right-hand input word testicle ‘a typically paired male reproductive gland that produces sperm and secretes testosterone […]’ (MWD), as well as the fact that the left-hand member bastard refers to ‘a contemptible, objectionable person’, the blend as a whole lends itself to a convincing analysis as semantically left- headed. On the other hand, the nominal blend cashish meaning ‘money’ requires a somewhat closer semantic analysis.47 Although the blend is not a hyponym of either of Lexis, 14 | 2019 157 the two input words, but rather a hypernym of the left-hand noun, it still receives a semantically left-headed interpretation, with the right-hand member hashish (seen ‘as a desirable/valuable commodity’) serving an emphasizing function. An interesting example is the blend Crentley which is used in US black slang somewhat humorously to designate ‘a Chrysler 300 automobile that looks like a fake–ss Bentley’. Accordingly, it is the left-hand member Chrysler that is the semantic head proper of the blend. In other words, Crentley is a type of Chrysler automobile which at the same time happens to resemble a Bentley, perhaps due to some adjustments made to the car. The synonymous adjectival blends craptastic and craptageous ‘(US) exceptionally mediocre’ are also considered left-headed because their meaning primarily depends on the almost identical meaning of the left-hand input word, i.e. the vulgar slang adjective crap ‘extremely poor in quality’ (OED).48 Their respective right-hand input words, fantastic and the presumed advantageous appear to be utilized here for emphasizing and ironic effects.49 In the blend wegro, used to denote ‘(US black) a white person who takes on black culture and style’, it is quite obvious that the left-hand input word white ‘(also White) A member of a light-skinned people, especially one of European extraction’ (OED) is semantically dominant. It is also noteworthy that the right-hand member negro itself involves some sort of meaning specialization here, since it is used in reference not only to ‘a member of a dark-skinned group of peoples […]’ (OED), but more specifically to their lifestyle. The same kind of interpretation seems to be relevant in the case of the blends gertoss meaning ‘(N.Z. teen) a young woman’, hunty which is used as ‘(US gay) a term of supposed endearment, with sarcastic overtones’, pootenanny meaning ‘(US black) the vagina’, and scrav ‘(UK juv.) to borrow or steal (usu. money)’. The decision to list hunty here is due to the fact that it is still considered an endearment, although its primarily positive connotation is no longer that prominent, but somewhat negatively modified by the noun cunty, especially outside gay community. The meanings of the input words of the blend pootenanny ‘the vagina’, namely punaany ‘the female genitals’ and hootenanny ‘a general term of abuse’ clearly indicate that the left-hand member is the semantic head. In the nominal blend stray denoting ‘a heterosexual man with homosexual tendencies’, the noun ‘person’ must be added to interpret it as semantically left-headed. As regards the semantic headedness of the verb scrav, although GDoS states that its left-hand input word scavenge is to be understood as belonging to Standard English, it seems more reasonable to consider it part of slang semantics, where the meaning of scavenge is ‘(Aus.) to pilfer’. This way, the blend as a whole is synonymous with its left-hand input word and thus taken to be semantically left-headed. By definition, double-headed or semantically coordinate blends include two members that carry equal semantic weight [Mattiello 2008: 89; 2013: 25]. In my data set these include: Blaxican, bonerific, cadazy, chillax, chunt, cramazing, craptard, datty, faburrific, hangry, knack, neek (sense 1), nugger, requestion, shart, shwasted, slore, smim, spleefer, spuzz, twasted, Vaalie, and widdle. However, among these blends, some appear to be closer to what I would term ‘true coordinate blends’, i.e. the blends in which a hybrid form imitates a hybrid concept [Thornton 1993: 148, 150–151; Renner 2019: 38]. Such examples are: Blaxican ‘(US black) one who is of mixed black and Mexican blood’, craptard ‘(US) one who is both foolish and an advocate of stupidity’, hangry ‘(teen) hungry and angry’, neek (sense 1) ‘a dull or unpopular person, esp. one who is interested in technology’ (CED), nugger50 ‘(US) an African American quadriplegic’, Lexis, 14 | 2019 158 requestion ‘(teen) a request and a question’51, shart ‘to expel faecal matter unintentionally when breaking wind’, smim ‘(UK juv.) one who is both highly conformist and physically unco-ordinated’, Vaalie ‘(S.Afr.) a native of the Transvaal, considered a peasant, a rustic, an unsophisticated person and generally looked down upon by the citizens of Cape Town, esp. when they appear there on holiday’.52 Unlike them, the blends such as: bonerific53, cadazy, chillax, chunt, cramazing54, datty, faburrific, knack, shwasted, slore, spleefer, spuzz, twasted, and widdle all represent combinations of two input words which happen to be (near-)synonyms.55 Consider, for instance, the noun slore in which the derogatory terms slut and whore are fused to produce an equally disparaging meaning, i.e. ‘(US campus) a derog. term for a sexually active woman’, or the noun widdle which is used to mean exactly the same as its two informal input words (wee and piddle), i.e. ‘(mainly juv.) an act of urination’.56 Consequently, blends like these are often considered “mere semantic alternatives of existing words” [Mattiello 2008: 24, 161; see also Green [2016: 12] for slang being composed of “novel-sounding synonyms (and near synonyms) for standard words and phrases”]. Furthermore, such blends seem to reflect the fact that “synonymy blocking is overridden by individual creativity, and appears to be irrelevant in clips […] and slang […]” [Miller 2014: 30]. This being the case, the following question arises: “Why do word blends with near-synonymous composites exist and persist?” [Evans & Steptoe- Warren 2015]. One possible explanation for the existence and persistence of such blend words in language is their “use in different sentence constructions”, as well as “very subtle semantic differences or identity implicature” [Evans & Steptoe-Warren 2015: 19, 25–26]. Finally, the exocentrics or those (slang) blends in which the semantic head is outside the blend [Mattiello 2008: 24; Bauer et al. 2013: 485] constitute a visible minority in my data set, since I identified only 6 such examples. These slang blends are used primarily to refer to people or their body parts, inanimate objects, and some abstract concepts such as: bromance, Koreegro, mangina, neek (sense 2), schmiddy, and yestergay. Consider first the blend bromance which is used to denote ‘a non-sexual relationship between males’. Namely, its second member (SE romance) is highly unlikely to represent the semantic head, since bromance is definitely not a kind of romance. That is, it seems rather difficult to treat it in some figurative manner or generalize the meaning of romance to the extent it excludes a feeling of romantic love that happens to be inherent in it (cf. the definitions of romance in the OED or MWD). Similarly, Koreegro is neither Korea nor a negro, but ‘(US black) an Asian person pretending to be a gangasta’. An alternative endocentric (i.e. left-headed) reading of Koreegro, whereby a transfer of metonymic, or more precisely synechdochic (pars pro toto) meaning occurs, seems less plausible, however. Put differently, it is rather difficult to conceive of a criterion by which Korea, and not, for instance, Japan or China, is representative of Asia or Asian people. Thus, one cannot but ask how entrenched figurative meanings have to be in language to account for an endocentric reading in cases such as the previous two (cf. Halupka-Rešetar & Lalić-Krstin [2009: 121]). The blend yestergay appears to present the hearer/reader with a similar interpretation problem, since it is neither (a hyponym of) yesterday, nor (of) gay, but ‘a former homosexual man who has (re-)adopted heterosexuality’. Also, no dictionary consulted for the purposes of this paper provides yesterday with a figurative (adjectival) meaning of ‘former’, which (if existed) would probably lend the blend as a whole to a fairly simple endocentric analysis, i.e. ‘an ex- gay’. However, although the noun gay “in its modern sense typically refers to men […] Lexis, 14 | 2019 159 in some contexts it can be used of both men and women” (see the OED’s entry for gay). That is to say, even if there were such a figurative (adjectival) meaning of yesterday, one might also come to interpret yestergay as a former homosexual woman (who is not necessarily in a heterosexual relationship). On the basis of the evidence presented, yestergay is perhaps best described as exocentric. The other three examples, namely mangina ‘(gay) anus’, neek ‘(UK teen) a drug dealer’, and schmiddy ‘(Aus.) a beer glass that is smaller than a schooner but larger than a middy’57 are, however, far more representative examples of exocentric blends. Conclusion On the basis of the above analysis of some of the slang blends in contemporary English, several conclusions can be reached (see also Table 1 below for a summary of the parameters used in the analysis of every single unit). Somewhat surprisingly, although the patterns used in the formation of these slang blends are quite diverse, almost half of them rise through what is considered to be a preference among English blends in general, i.e. by merging the initial part of the first word and the final part of the second one. This diversity of blending patterns is mainly due to the violation of many rules of morphological grammar, which is why blends are disfavored from the domain of Natural Morphology (NM) and more readily described as manifestations of EM. In addition, most of these 60 creations are nouns, regardless of the word classes of their inputs, with only few adjectives and verbs attested. With regard to the blends’ morphotactics, half of them (see Section 3.1. (a), (d) (cadazy and scrav), (f), and (i)) are formed from two shortened rather than whole words, which results in their greater morphotactic opacity and is perhaps best explained by the very register (i.e. slang) they are created in. If the semi-complete blends are added to this number, then it seems safe to say that there is an inclination for the obscuration of blend words in slang. That is to say, certain blending formation patterns serve quite well the communicative functions of slang, “in which outsiders of a group are not meant to understand what is communicated to insiders” [Ronneberger-Sibold 2010: 206; but cf. also Mattiello 2008: 15]. Interpretation in such cases is greatly facilitated by an overlap, which is identified in as many as 28 slang blends. Namely, 13 of these blends manifest a phonemic-cum-graphemic overlap, 9 blends manifest a phonemic overlap, whereas a graphemic overlap is identified in only 6 blends (cf. Renner [2019: 34] for similar results). Regarding the morphosyntactic headedness of the slang blends, in over half of them it is the right-hand input word that transfers its grammatical properties to the output and is thus considered grammatically dominant. On the other hand, the preference of the slang blends for the semantic right-headedness is not that salient, since only 21 out of 61 examples can be taken to be right-headed, which may also be attributed to the cryptic and elusive character of slang. Furthermore, a frequent use of figurative (slang) meanings of the input words makes the interpretation of these blends cognitively more demanding. It should also be borne in mind that, although I have classified all the slang blends from my data set into either endocentric or exocentric in nature, not all of them represent (proto)typical examples of endocentricity and exocentricity. Thus, it is perhaps more appropriate to conceive of the semantic headedness of these blends as a continuum of cases, with endocentric slang blends in which the head is immediately Lexis, 14 | 2019 160 identifiable, as in budiquette, dopalicious, or cashish at one end of the continuum and exocentric slang blends in which the head is not identifiable from either of the input words, as in neek (sense 2) or Koreegro at the other [Halupka-Rešetar & Lalić-Krstin 2009: 121–122]. In between the two ends, but closer to the endocentrics are those blends whose heads acquire this status via metaphoric and metonymic transfers of meaning (e.g. blamps) [Bauer 2008].58 Another observation that might be said to complement a rather peculiar nature of these slang blends concerns the length (in terms of syllables) of the semantic head. Namely, only 15 out of 32 endocentrics have the longer of the two input words as the semantic head. These are: blamps, Blinglish, budiquette, buffugly, darty, Datto, dopalicious, friendscape, Jailic, jerkitude, kidult, mackadelic, scrav, shegarry, and twoonie. The ways in which the non-head modifies the semantic head in these slang blends are manifold. Some of the possible readings of the modifiers are: quality (e.g. blamps, buffugly, cashish, dopalicious, fratastic, jerkitude, etc.), quantity or worth (e.g. twoonie), concerned with (e.g. budiquette), membership (e.g. sororowhore, Datto), time (e.g. darty), location (e.g. Jailic), the instrument (e.g. slock), subject (e.g. grape), object (e.g. friendscape, gertoss, spum). There is also a small number of the slang blends, such as basticle, craptastic, craptageous, shegarry, etc. in which the modifier-head relationship is perhaps best described as the one where the modifier acts as an amplifier. Concerning the already mentioned observation that many of the slangy formations here represent no more than semantic alternatives to the existing standard words and phrases, they nonetheless increase the potential of a language, as well as the area of word-formation research [Mattiello 2005: 18]. With regard to the relationship between the morphosyntactic and the semantic head(s) of the slang blends, in more than half of them the two types of head coincide. I consider the two types of head to coincide if the blend has two morphosyntactic and two semantic heads, as in: chillax, chunt, datty, hangry, knack, neek (sense 1), nugger , requestion , shwasted, slore , smim , spleefer , spuzz , twasted, Vaalie, and widdle. Table 1: Parameters used in the morphological and semantic analysis of contemporary English slang blends Number of blends in Parameter the data set Formation the first part of w159 + the second part of w260, with a 26 pattern possible overlap w1 + the second part of w2, with a possible overlap 14 the first part of w1 + w2, with a possible overlap 9 (the initial or final part of) w2 is inserted within the 4 part of w1 w1 + w2, with an obligatory overlap 3 the first part of w1 + the first part of w2 1 w1 + a medial part of w2 1 Lexis, 14 | 2019 161 the first part of w2 is inserted within w1 1 the second part of w1 + the second part of w2 1 Overlapping 28 Non-overlapping 32 right-headedness 33 left-headedness 6 Morphosyntactic headedness double- 21 headedness right-headedness 21 left-headedness 11 Endocentricity Semantic headedness double- 23 headedness Exocentricity 6 Last but not least, I hope some of the above conclusions have shed more light on the issue of headedness in slang blends, as well as on the variety of mechanisms that are exploited in the creation of blends in this specific register. It is also my hope that this contribution will act as a stimulus for further research into extra-grammatical phenomena (in different registers), and especially blending as a popular way of adding new words to the English lexicon. BIBLIOGRAPHY ADAMS Valerie, 1973, An Introduction to Modern English Word-Formation, London/New York: Longman. ARONOFF Mark, 1976, Word Formation in Generative Grammar, Cambridge: MIT Press. BALTEIRO Isabel, 2013, “Blending in English Charactoons”, English Studies 94 (8), London and New York: Routledge, 883–907. BAREŠ Karel, 1974, “Unconventional Word-Forming Patterns in Present-Day English”, Philologica Pragensia, 17, Praha: Academia, nakladatelství ČSAV, 173–186. BAT-EL Outi, 2006, Blend, in BROWN Keith (Ed.), Encyclopedia of Language & Linguistics, Second Edition, Vol. 2, Oxford: Elsevier, 66–70. Lexis, 14 | 2019 162 BAUER Laurie, 1983, English Word-Formation, Cambridge: Cambridge University Press. BAUER Laurie, 1990, “Be-Heading the Word”, Journal of Linguistics, 26 (1), Cambridge: Cambridge University Press, 1–31. BAUER Laurie, 2003 [1988], Introducing Linguistic Morphology, 2nd edition. Edinburgh: Edinburgh University Press. BAUER Laurie, 2008, “Exocentric Compounds”, Morphology, 18 (1), Dordrecht: Springer, 51–74. BAUER Laurie, LIEBER Rochelle & PLAG Ingo, 2013, The Oxford Reference Guide to English Morphology, Oxford: Oxford University Press. DRESSLER Wolfgang U., 2000, “Extragrammatical vs. Marginal Morphology”, in DOLESCHAL Ursula & THORNTON Anna M. (Eds.), Extragrammatical and Marginal Morphology, München: Lincom Europa, 1– 10. DRESSLER Wolfgang U., 2005, “Word-Formation in Natural Morphology”, in ŠTEKAUER Pavol & LIEBER Rochelle (Eds.), Studies in Natural Language and Linguistic Theory Handbook of Word-Formation, Dordrecht: Springer, 267–284. EVANS Thomas Rhys & STEPTOE-WARREN Gail, 2015, “Why Do Word Blends with Near-Synonymous Composites Exist and Persist? The Case of Guesstimate, Chillax, Ginormous and Confuzzled”, Psychology of Language and Communication, 19 (1), Berlin: De Gruyter, 19–28. GREEN Jonathon, 2016, Slang: A Very Short Introduction, Oxford: Oxford University Press. GRIES Stefan Th., 2012, “Quantitative Corpus Data on Blend Formation: Psycho- and Cognitive- Linguistic Perspectives”, in RENNER Vincent, MANIEZ Francois & ARNAUD Pierre J. L. (eds.), Cross- Disciplinary Perspectives on Lexical Blending, Berlin: Mouton de Gruyter, 145–167. HALUPKA-REŠETAR Sabina & LALIĆ-KRSTIN Gordana, 2009, New blends in Serbian: Typological and headedness-related issues, in RUŽIĆ Vladislava & ŠLJUKIĆ Srđan (Eds.), Godišnjak Filozofskog fakulteta u Novom Sadu, XXXIV(1), Novi Sad: Filozofski fakultet, 115–124. HUDDLESTON Rodney D. & PULLUM Geoffrey K., 2002, The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press. KEMMER Suzanne, 2003, “Schemas and Lexical Blends”, Motivation in Language Current Issues in Linguistic Theory, Amsterdam/Philadelphia: John Benjamins, 69–97. KÖRTVÉLYESSY Lívia, 2014, “Evaluative Derivation”, in ROCHELLE Lieber & ŠTEKAUER Pavol (Eds.), The Oxford Handbook of Derivational Morphology, Oxford: Oxford University Press, 296–316. KUBOZONO Haruo, 1990, “Phonological constraints on blending in English as a case for phonology- morphology interface”, in BOOIJ Geert E. & VAN MARLE Jaap (Eds.), Yearbook of Morphology 1990, 3, Dordrecht: Kluwer, 1–20. LEHRER Adrienne, 2003, “Understanding Trendy Neologisms”, Italian Journal of Linguistics 15 (2), Pisa: Pacini Editore, 369–382. LEHRER Adrienne, 2007, “Blendalicious”, in MUNAT Judith (Ed.), Lexical Creativity, Texts and Contexts, Amsterdam/Philadelphia: John Benjamins, 115–133. MARCHAND Hans, 1969, The Categories and Types of Present-Day English Word-Formation: A Synchronic- Diachronic Approach, München: C.H. Becksche Verlagsbuchhandlung. Lexis, 14 | 2019 163 MATTIELLO Elisa, 2005, “The Pervasiveness of Slang in Standard and Non-Standard English”, in LONATI E. (Ed.), Mots Palabras Words: Studi Linguistici 6, https://www.ledonline.it/mpw/allegati/ mpw0506Mattiello.pdf MATTIELLO Elisa, 2008, An Introduction to English Slang: A Description of Its Morphology, Semantics and Sociology, Milano: Polimetrica. MATTIELLO Elisa, 2013, Extra-Grammatical Morphology in English: Abbreviations, Blends, Reduplicatives, and Related Phenomena, Berlin: De Gruyter Mouton. MILLER Gary, 2014, English Lexicogenesis, Oxford: Oxford University Press. MÜLLER Peter O., OHNHEISER Ingeborg, OLSEN Susan & REINER Franz, 2015, Word-Formation: An International Handbook of the Languages of Europe, Vol. 1, Berlin: De Gruyter Mouton. OLSEN Susan, 2014, “Delineating Derivation and Compounding”, in ROCHELLE Lieber & ŠTEKAUER Pavol (Eds.), The Oxford Handbook of Derivational Morphology, Oxford: Oxford University Press, 26– 49. PINKER Steven, 2007, The Stuff of Thought: Language as a Window into Human Nature, London: Penguin Books. PLAG Ingo, 2003, Word-Formation in English, New York: Cambridge University Press. POUND Louise, 1914, Blends: Their Relation to English Word Formation, Heidelberg: C. Winter. RALLI Angela & XYDOPOULOS George J., 2012, “Blend formation in Modern Greek”, in RENNER Vincent, MANIEZ Francois & ARNAUD Pierre J.L. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: Mouton de Gruyter, 35–50. RENNER Vincent, 2015, “Lexical Blending as Wordplay”, in ZIRKER Angelika & WINTER-FROEMEL Esme, Wordplay and Metalinguistic / Metadiscursive Reflection, Berlin: De Gruyter, 119–134. RENNER Vincent, 2019, “French and English Lexical Blends in Contrast”, Languages in Contrast. International Journal for Contrastive Linguistics 19 (1), Amsterdam/Philadelphia: John Benjamins, 27– 47. RONNEBERGER-SIBOLD Elke, 2006, “Lexical Blends: Functionally Tuning the Transparency of Complex Words”, Folia Linguistica, 40 (1–2), Berlin: Mouton de Gruyter, 155–181. RONNEBERGER-SIBOLD Elke, 2010, “Word Creation”, in RAINER Franz, DRESSLER Wolfgang U., KASTOVSKY Dieter & LUSCHÜTZKY Hans Christian (Eds.), “Variation and Change in Morphology: Selected Papers from the 13th International Morphology Meeting, Vienna, February 2008”, Amsterdam/Philadelphia: John Benjamins. RONNEBERGER-SIBOLD Elke, 2012, “Blending between grammar and universal cognitive principles”, in RENNER Vincent, MANIEZ Francois & ARNAUD Pierre J.L. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: Mouton de Gruyter, 115-144. SORNIG Karl, 1981, Lexical Innovation: A study of slang, colloquialisms and casual speech, Amsterdam: John Benjamins. THORNTON Anna M., 1993, “Italian Blends”, in TONELLI Livia & DRESSLER Wolfgang U., Natural Morphology: Perspectives for the Nineties, Padova: Unipress, 143–155. WILLIAMS Edwin, 1981, “On the Notions of ‘Lexically Related’ and ‘Head of a Word’”, Linguistic Inquiry, 12 (2), Cambridge: The MIT Press, 245–274, https://www.jstor.org/stable/4178218 Sources Lexis, 14 | 2019 164 “Jamaican Patois.” n.d. Wikipedia. Accessed June 14, 2019. https://en.wikipedia.org/wiki/ Jamaican_Patois “Patterns of Word Formation” n.d. Oxford Dictionaries. Accessed May 6, 2019. https:// en.oxforddictionaries.com/explore/patterns-of-word-formation/ CED. Collins English Dictionary. n.d. Accessed May 5, 2019. https://www.collinsdictionary.com/ dictionary/english/ GDoS. Green’s Dictionary of Slang. n.d. Accessed January 2019. https://greensdictofslang.com/ Google. https://www.google.com/ MWD. Merriam-Webster Dictionary. n.d. Accessed May 4, 2019. https://www.merriam- webster.com/ OED. Oxford English Dictionary. n.d. Accessed May 4, 2019. https://en.oxforddictionaries.com/ The Rice University Neologisms Database. Accessed June 14, 2019. https://neologisms.rice.edu/ UD. Urban Dictionary. n.d. Accessed May 27, 2019. https://www.urbandictionary.com/ NOTES 1. For a discussion on slang as a linguistic register, see Green [2016: 9–12]. 2. Marchand [1969: 451] also claims that blends do not have a grammatical, but a stylistic status only. 3. That “blends are generally interpreted in the same way that compounds are, though not necessarily in the same proportions” was also recognized by Bauer et al. [2013: 485]. 4. See Renner [2019: 36–37] for a discussion about the restrictions of the hyponymy test. 5. Regarding exocentric compounds, Williams [1981: 250, 261] claims “that they are derived by headless rules”. 6. Note that this claim seems to be in sharp contrast with what many authors categorize under exocentric blends (cf., for instance, Mattiello [2013: 124–125, 130]; Renner [2019: 37–38]). 7. But see Gries’s [2012: 154–155] classification of English blends into synonymic, co-hyponymic, contractive, frame relation, and other (e.g., antonymy, derivation, etc.), based on the semantic relation between the two source words. 8. Since GDoS is not designed to automatically extract products of blending and hence does not make use of the term ‘blend’, I had to decide on the working definition of a blend first and then manually extract the examples whose etymology meets the criteria in the above definition. It (GDoS) can, however, automatically extract the products of some other word-formation patterns such as abbreviation. It is also worth noting that there is sometimes a question mark used next to one (or rarely both) of the input words in GDoS, signalling that it is just a possible input word, but also confirming the statements made elsewhere in the relevant literature that in some blends it is extremely difficult or almost impossible to unambiguously determine their input words. There are also few cases in my data set in which I had to presume the second input word (e.g. craptastic, craptageous), since no etymology was provided. 9. Cf. Mattiello [2005; 2008] for the sources she used in her slang-related research. 10. According to Kemmer [2003: 80], “of all blends, overlap blends are the type that exhibits perhaps the most striking creativity of usage”. 11. Although the lemma in GDoS reads ‘basticles’, the example of its use (“Surround the sorry basticles and wipe them all out but do it from the perimeter of the base please.”), as well as its Lexis, 14 | 2019 165 definition (‘a general term of annoyance, abuse etc.’) seem to suggest it should be ‘basticle’ instead (cf. also the entry in the UD). 12. Standard English. 13. Note that the double -r- from terrific has been reduced to only one -r- in bonerific. 14. Here the blend coiner decided to employ the phonological rather than the orthographic form of the first fragment. The reason behind such decision may be the fact that slang “is predominantly associated with spoken language” [Mattiello 2008: 49]. 15. Instead of retaining the vowel -e- from dope or delicious, which would result in exactly the same pronunciation with a schwa, the two fragments are blended by means of the vowel -a- (cf. Mattiello [2008: 140]). 16. The addition of -k- to the resultant combination of the two fragments does not influence the phonological form of the blend, but it seems to indicate the tendency of slang creators to make its lexis more obscure. 17. This blend, as well as some other (e.g. Crentley, spleefer, spum, stray, wanksta, wegro, and widdle) seem to suggest the tendency of slang blends to rhyme with the second input word. 18. The insertion of -u- also complicates the recoverability of the input words. 19. It seems not to be unusual for a slang blend to represent a homograph and/or a homophone of the standard term (but cf. Aronoff [1976: 43]). See also grape below. 20. Yet another blend in the creation of which the phonetic rendering (of the word gangsta) is employed (cf. also the entry for wanksta in the CED). 21. The blend may as well be interpreted as an example of fragment blend, i.e. in such a way that the fragment bla- is blended with the fragment -xican. My decision to consider it the result of the second rather than the first formation pattern is influenced by the already mentioned slang’s identification with speech primarily (cf. footnote 14). 22. See footnote 8. 23. Although it says in GDoS that friendscape is a noun, the definition and the example of its use clearly indicate it should be labelled as a verb. 24. The word Gaelic can also be pronounced with a schwa, in which case the overlap would be reduced to the phoneme or the letter -l-. 25. Although the etymology of mackadelic in GDoS reads ‘[mack n.2 (1) + SE sfx -delic]’, it seems to me that the formation of mackadelic here crucially depends on the word phychedelic, both morphologically and semantically [cf. Mattiello 2008: 21 for the interpretation of shagadelic]. Considering the semantics of mackadelic ‘self-promoting, parading the qualities of a pimp’, but also the fact that pimps are known for their flashy attire, the adjective psychedelic is most probably used with the meaning ‘having intense, vivid colours or a swirling abstract pattern’ (OED). 26. Not only is the consonant -f- doubled, but also the second input word fugly represents a blend itself, which makes this neoformation all the more complex and difficult to interpret. 27. The output word is further obscured by the substitution of the letter -i- with the letter -e-. The alteration of the spelling of girl may in fact be attributed to the proclivity of slang creators to manipulate word pronunciations [Mattiello 2008: 41]. 28. Standard Australian English. 29. Although its etymology in GDoS reads ‘[? SE spastic + mimic]’, there seems to be no reason why spastic should not be interpreted as part of slang lexis, where it is used to mean ‘(offensive, slang) a clumsy, incapable, or incompetent person’ (CED), especially if one considers the meaning of the blend as a whole ‘(UK juv.) one who is both highly conformist and physically unco- ordinated’. Also, since no part of speech labelling is provided for the inputs, I take mimic to be a noun too, in the sense of ‘a person who adopts the attitudes, behaviour, dress, etc of the group to which he or she belongs’ (cf. the entry for conformist in the CED), or of one who simply mimics what other people do. Lexis, 14 | 2019 166 30. E.g. “Salman! Don’t be a basticle!” (UD). 31. (In)animacy, abstractness, and similar features are considered here to be syntactic rather than semantic or, in the words of Bauer [1990: 24], syntactic features based in semantics. 32. E.g. “Oh, hunty, let me tell you something.” Cf. also the three entries for ‘hunty’ in the Rice University Neologisms Database, available at: https://neologisms.rice.edu/ 33. Due to a rather limited number of examples of the word’s use in GDoS, the suggestion made here is rather tentative. 34. E.g. “I ended up chillaxing at Mike’s place last night.” 35. E.g. “When a girl has a bonerific body, but then a face that would make Mother Theresa cry for mercy.”; “Woah, that chick was bonerific!” (UD). 36. E.g. “Hot as in bonerifically grrrrrrrrrrrrreat!” (Google). Also, when I entered ‘bonarooly’ on Google (using double quotation marks), I did not find any examples of its use. 37. This blend seems to have firmly established itself in some of the leading dictionaries of General English (e.g. CED). 38. “Jamaican Patois […] is an English-based creole language […] spoken primarily in Jamaica and the Jamaican diaspora […]”, available at: https://en.wikipedia.org/wiki/Jamaican_Patois 39. E.g. “He has no budiquette, he completely skipped me in the rotation.” (Google). 40. E.g. “We’re going to black out while it’s light out at the darty!” 41. E.g. “Stankonia packs more dopalicious jams onto one album than any hip-hop (or not) album in a long time.” It is noteworthy that delicious “is used in American slang to form adjectives with the meaning ‘embodying the qualities denoted or implied by the first element to a delightful or attractive degree’” [Mattiello 2008: 121]. 42. E.g. “He’s so fratastic in his silk bowtie and pastel shorts […].” 43. E.g. “Those on the H-Blocks created ‘Jailic’ to counter the Crown’s repression.” 44. E.g. “Someone woulda put a couple of padlocks in a sock and slocked your dome.” 45. E.g. “I’ve ’ad nuff of dese shegarries!” 46. E.g. “Sorry, I accidentally deleted you when I was friendscaping.” 47. E.g. “In the safe they found [...] bundles of cashish – pounds, euros and dollars.” 48. E.g. “Readers suggested bad products henceforth be described as: craptastic, craptageous, […].” 49. Note also that this ironic use of the adjective fantastic (in blends) seems not to be uncommon [https://en.oxforddictionaries.com/explore/patterns-of-word-formation/]. On the other hand, the very few blends containing the fragment –tageous (e.g. fun-tageous or Eventageous (Google)) show no such tendency in the case of advantageous. 50. Its left-hand member nugget is used in US slang to mean ‘a quadriplegic’. 51. It is basically a request made in the form of a question (cf. UD). 52. This slightly modified definition is taken from GDoS. 53. E.g. “Beautiful Cameron Diaz Looks Bonerific In Her Redhead Version” (Google). 54. According to GDoS, the adjective crazy is used in the blend with the meaning ‘(US black/ beatnik) a general intensifier, wonderful, amazing, weird, bizarre, according to context.’. 55. Note, however, that if one adopts the semantic characterization of blends as proposed in Bauer et al. [2013: 483–485], both these groups belong to what they call “compromise coordinative blends” (e.g. avoision, puggle). 56. E.g. “Riddle of where to have a widdle” (Google). 57. E.g. “he thought the schmiddy – a 355ml-size glass, between the middy and the schooner – was ‘overpriced’”. 58. Note that Halupka-Rešetar & Lalić-Krstin [2009: 121–122] suggest that blends which involve metaphoric transfer of meaning are “more readily classified as endocentric” than those metonymy-based blends. 59. The first word. Lexis, 14 | 2019 167 60. The second word. ABSTRACTS The present paper aims at investigating both morphosyntactic and semantic headedness of 60 (61 senses) contemporary English slang blends, as well as some of the blends’ formal properties, namely the patterns by which they are formed and the syntactic categories they and their input words belong to. The blends whose first use was recorded between 2000 and 2019 are excerpted from the online version of Green’s Dictionary of Slang. The reason why slanguage is chosen for blend excerption is the fact that slangy formations are conscious rather than spontaneous [Mattiello 2008: 16], the same way that blends are, and also because slang, like blending, has a tendency for clipping words [Mattiello 2008: 141]. Among 9 formation patterns identified, half of the blends are formed from two shortened rather than whole words, which results in their greater morphotactic opacity. A qualitative and quantitative analysis of morphosyntactic headedness of the slang blends shows that in most cases it is the right-hand member that functions as the head. On the other hand, a qualitative and quantitative analysis of semantic headedness shows that the preference of the slang blends for the semantic right-headedness is not that prominent, since only 21 out of 61 examples can be said to be semantically right-headed, which may in part be explained by the cryptic character of the slang register. Also, only 15 out of 32 endocentric slang blends appear to have the longer of the two words as the semantic head. Finally, although the slang blends analyzed here are for the most part no more than the semantic alternatives to the existing standard words and phrases, they nonetheless increase the potential of a language, as well as the area of word-formation research [Mattiello 2005: 18]. L’objectif de ce travail est d’examiner la centricité morphosynthaxique et sémantique de 60 mots-valises (61 sens) dans l’argot de l’anglais moderne, ainsi que certaines de leurs caractéristiques fonctionnelles, telles que les modèles selon lesquels ils ont été créés, et le type de mots auxquels ils appartiennent, ainsi que leurs constituant. Le corpus de recherche est constitué de mots apparus entre 2000 et 2019, extraits de l’édition électronique du dictionnaire Green’s Dictionary of Slang. La raison pour laquelle l’argot a été choisi comme source du corpus est que les termes utilisés en argot, tout comme les mots-valises, sont des créations délibérées [Mattiello 2008 : 16], mais aussi parce que dans l’argot, de même que dans la fusion des mots-valises, existe une tendance à raccourcir les mots[Mattiello 2008 : 141]. Ainsi, en analysant les aspects structurels de 60 mots-valises, nous avons identifié 9 types de formations différentes. Dans la moitié des cas, le mot-valise est formé par fusion de deux mots abrégés, ce qui a pour conséquence une transparence morphotactique plus faible. Une analyse qualitative et quantitative de la centricité morphosynthaxique dans nos mots-valises montre que, dans la plupart des cas, le terme de droite est un centre morphosyntaxique. D’autre part, l’analyse qualitative et quantitative de la centricité sémantique dans nos mots-valises a révélé que leurs créateurs n’ont pas tendance à placer le centre sémantique dans le terme de droite, car seuls 21 exemples sur 61 ont un centre sémantique dans le terme de droite, ce qui, du moins en partie, peut s’explique par le mystère du registre dans lequel ces mots-valises sont formés, à savoir l’argot. De plus, dans seulement 15 des 32 mots-valises endocentriques du corpus, constitués de plus de deux mots, le terme central est simultanément le centre sémantique du mot-valise. En fin de compte, bien que les mots-valises de l’argot analysés ici sont principalement des synonymes Lexis, 14 | 2019 168 de mots et d’expressions standard existants, ils contribuent au développement du potentiel d’une langue, en l’occurrence de l’anglais, ainsi qu’au développement de la recherche dans le domaine de la formation de mots [Mattiello 2005 : 18]. INDEX Keywords: blends, slang, contemporary English, morphosyntactic headedness, semantic headedness, formal characteristics Mots-clés: mots-valises, argot, langue anglaise moderne, centricité morphosyntaxique, centricité sémantique, caractéristiques formelles des mots-valises AUTHOR GORICA TOMIĆ University of Kragujevac, Serbia Faculty of Philology and Arts Center for Language and Literature Research [email protected] Lexis, 14 | 2019 169 New lexical blends in The Simpsons: a formal analysis of English nonce formations and their French translations Adam Renwick and Vincent Renner Introduction 1 New words are not always coined with the purpose of becoming institutionalized, i.e. of becoming part of the lexicon of a community of speakers. As Hohenhaus [2005: 365] puts it, “nonce can be the first stage in a longer life-span of a word but need not be – and mostly it is also the last stage”. New words serve a variety of communicative purposes and a cline of likelihood of institutionalization may be posited, where the playful creation in (1) could be placed at one extreme (institutionalization is relatively very unlikely) and the terminological creation in (2) at the other extreme (institutionalization is relatively very likely): (1) purr + perplexing > purrplexing (The Simpsons, Season 26, Episode 18) (2) adsorbed + atom > adatom (physical chemistry)1 2 Studying lexical blends should be especially enlightening in this regard as a correlation between functional type and formal variation may be postulated. Our hypothesis is that playful nonce formations might be noticeably different in their formal characteristics than institutionalized blends as a class because the identification of the source elements and the construal of meaning can only take place online, i.e. during the actual perception of speech. By definition, the various source elements of a blend do not all appear in full and the operation of blending leads to various degrees of formal opacification. Haplologic blends, which are coined through segment overlap, as illustrated in (3), can be considered to be minimally opaque while outputs resulting Lexis, 14 | 2019 170 from the clipping of polysyllabic elements to mono- or infra-syllabic constituents, as illustrated in (4), are maximally opaque: (3a) hip-hop + opera > hip-hopera (3b) winter + interim > winterim (4a) columbite + tantalite > coltan (4b) binary + digit > bit 3 Playful nonce formations are thus expected to be characterized by a lesser degree of formal opacity than institutionalized items. This would be in line with previous research by Ronneberger-Sibold [2006], who reports in her study of the relative transparency of blend types in German that 66% of the 612 units in her dataset of literary and journalistic nonce formations, but only 16% of the 220 units in her dataset of brand names, are maximally transparent, i.e. coined through haplologic blending or the full (phonemic) overlap of source elements (as in Jewbilee, from Jew and jubilee). 4 In order to assemble a large dataset of playful nonce blends in present-day English, we resorted to collecting items from a tailor-made corpus comprising 29 seasons of scripts from the US animated TV series The Simpsons, a long-running sitcom identified for its lexical creativity and expected to contain a sizable number of blends. As directly comparable data were available in Hexagonal French in the form of translated script, a lexical analysis of the translations of the English nonce blends into French was also subsequently carried out. Analytic data on English and French institutionalized blends taken from Renner [2019] were used as points of comparison. 1. Methodology 1.1. English data 5 To collect the novel lexical blends appearing in The Simpsons, we first compiled an electronic corpus of transcripts of the 639 episodes from the 29 complete seasons broadcast up to mid-2018. This corpus was based on fan-created transcripts published to a website dedicated to transcripts of television shows, http:// transcripts.foreverdreaming.org/. It is of note that these transcripts do not necessarily exactly reflect the scripts used by the cast of the TV show when recording dialogue for the episodes and are subject to some misinterpretations as well as transcription and typographical errors. However, the consultation of a second website containing transcripts of the series (https://www.springfieldspringfield.co.uk/) and the examination of specific clips from the show available on various video sharing platforms guarantee that the corrected transcripts that were compiled are highly accurate records of the lexical content of the show. HTML coding was stripped to give raw text files for each season which were analyzed with the MonoConc Pro concordancer. These 29 files comprised some 1.5 million words (approximately 2,300 words per episode on average). 6 Given our focus on novel blends, we chose to proceed by the use of an exclusion corpus to remove already attested words. To create the exclusion corpus, we combined several existing stoplists of English words in order to limit the number of candidate forms. The first two stoplists were based on Webster’s Second International Dictionary in the form of wordlists named web2 (235,887 items) and web2a (76,206 items) which are included in Lexis, 14 | 2019 171 Unix/Linux-based operating systems. They encompass the large percentage of the English lexicon that has remained formally unchanged since the publication of this dictionary in 1934. They were supplemented by three further stoplists that include lexical innovations up to the first decade of the twenty-first century: the list of the 20,000 most common words in the English GoogleBooks corpus (https://github.com/ first20hours/google-10000-english/blob/master/20k.txt), the list of the American English spelling variants of the 10,000 most common words in the corpus (https:// github.com/first20hours/google-10000-english/blob/master/google-10000-english- usa.txt) and a customized list combining the wordlists developed by the SCOWL (And Friends) project (http://wordlist.aspell.net/). These five stoplists were merged to form a final exclusion corpus containing 402,933 words. Using our concordancer, we generated a list of all items used in the 29 seasons of The Simpsons that were not present in the exclusion corpus. This produced a set of 19,709 candidate forms, including 14,812 hapaxes. 7 These remaining candidates were then examined individually and discarded if they were simplex words or had been constructed through other word-formation processes like affixation or compounding. The boundaries of the concept of lexical blending have been subject to debate in the morphological literature (see e.g. López-Rúa [2004], Bauer [2012: 19-21], Beliaeva [2014: 45-47] and Renner [2015: 99-105] for a discussion). To make valid comparisons with institutionalized blends, we chose to adopt the definition of blending provided by Renner [2019: 29], namely that a blend is a constructed word which does not contain all of its source elements in full while satisfying the following three requirements: firstly, it does not contain a recurring word fragment which could be assimilated to a combining form (as in slumpflation, which can be considered a compound of slump and ‑ flation, or Margealicious, a Simpsonian coinage that was analyzed as the compounding of Marge and ‑licious with the help of a linking vowel); secondly, it does not manifest single external shortening (as in the case of blog, which can be considered a clipped form of weblog rather than a blend); and, thirdly, it is not coined through the clipping of the initials of a majority of its source elements, as is the case of zineb (< zinc + ethylene + bisdithiocarbamate), because, from a prototype-based perspective, such a formation is closer to the class of initialisms than to that of blends. This final sifting stage led to the identification of a total of 237 blends in a corpus of about 1.5 million words, at an average of 158 blends occurring per million words. 1.2. French data 8 To examine the manner in which the English blends were rendered in French, we created a corpus based on the Hexagonal French version of The Simpsons along the same lines as for the English corpus. We gathered the existing transcripts from fan- submitted transcripts of the first 26 seasons (www.simpsonspark.com/scripts.php), as those episodes in Seasons 27-29 had not yet been broadcast in Hexagonal French, meaning we are only able to offer a translational perspective on 217 novel English blends amongst the 237 analyzed. From the French transcripts, we then extracted the utterances corresponding to the 217 English blends. In cases where it was uncertain based from the transcript alone that the utterance in question contained a blend, we viewed the relevant part of the episode in question to determine the accuracy of the transcription as compared to the audio and to gather any further non-verbal information present onscreen before, during and following the utterance in question. Lexis, 14 | 2019 172 This method sufficed to determine whether a blend was indeed present, as well as the source elements used to form it, and it was also used in the few cases where no transcript was yet online for certain episodes in Season 26. 2. A formal analysis of English nonce blends 9 The dataset of English nonce blends comprises 237 items, including 4 three-element items (e.g. Frightmarestein (< fright + nightmare + Frankenstein)) which are not analyzed further in this section because of their marginal status. All blends are given in their orthographic form for simplicity’s sake, but they were retrieved from spoken data and the formal analysis below is thus solely based on the phonological form of each item. 2.1. Lexical shortening 10 The distribution of the various types of lexical shortening for the 233 two-element blends of the Simpsons dataset is presented in Table 1 and contrasted with that available for institutionalized blends (Renner [2019: 33]). Table 1: Distribution of English blends according to the type of lexical shortening Type of lexical shortening Institutionalized blends Nonce blends left-hand-side inner shortening 24% 17% (40 items) (e.g. Viagrogaine < Viagr[a] + Rogaine) right-hand-side inner shortening 21% 55.5% (129 items) (e.g. fudgesicle < fudge + [pop]sicle) double inner shortening 31% 11% (26 items) (e.g. lupper < lu[nch] + [s]upper) double right-shortening 14% – (e.g. cyborg < cyb[ernetic] + org[anism]) double left-shortening – 0.5% (1 item) (e.g. cueabunga < [barbe]cue + [cowa]bunga) haplologic blending 7% 12.5% (29 items) (e.g. galgebra < g[al + al]gebra) sandwich blending 1% 3.5% (8 items) (e.g. Rastafrogian < Rastaf[ar]ian + frog) other 2% – (e.g. ziram < zi[nc] + [ca]r[b]am[ate]) 11 The first striking difference between the two distributions is the absence of double right-shortened units like cyborg (< cybernetic + organism) or perfin (< perforated + initial) Lexis, 14 | 2019 173 in the set of nonce blends, highlighting that such items form a very atypical class of blends and are sometimes not even classified as blends, but as “clipping compounds” or “complex clippings”, because of their distinctive formal features (see e.g. Beliaeva [2019, Section 3.2.] and Renner [2019: 44] for a discussion). The dispreference for the pattern of double right-shortening in nonce blends may be explained in terms of (non-)recognizability of the source elements, each of them being canonically clipped to a monosyllable, which often causes the syllabic contour of either source element to be lost. 12 The other four non-marginal types of lexical shortening are found in the two sets of blends, but in a starkly different frequency order. The most notable difference lies in the overwhelming preference among nonce blends for the pattern of right-hand-side inner shortening, as illustrated in fudgesicle (< fudge + popsicle) and mathnasium (< math + gymnasium), which accounts for more than half of all items (55.5%). Correlatively, the proportion of double inner-shortened blends like lupper (< lunch + supper) and Purgatraz (< purgatory + Alcatraz) is dramatically lower in the nonce-blend set (11%). This underscores that the preference for the pattern of double inner shortening may not be universal, as is commonly believed (Beliaeva [2019, Section 3.3.]): One obvious regularity that is postulated in the literature as a defining feature of blends is that most blends combine the initial part of one word with the final part of another. 13 This relative dispreference for double inner shortening may, here too, be explained in terms of relative recognizability: an output which has retained one source element in full is formally more transparent than an output made of two fragments. This explanation may also be used to account for the relatively high percentage of haplologic blends like galgebra (< gal + algebra) and Gaybraham (< gay + Abraham) in the set of nonce blends. As pointed out in the Introduction, the blends in this class are minimally opaque, i.e. display the highest possible degree of recognizability. 2.2. Segment overlap 14 Segment overlap is not a defining, but a prototypical feature of lexical blending. It typically appears word-medially, around the point of splicing of the source elements or word fragments, as illustrated by the nonce blends in (5): (5a) bland + tandoori > bl[and + and]oori > blandoori (5b) Lamborghini + Bugatti > Lambor[g + g]atti > Lamborgatti 15 Identical segments may also appear non-medially in the two source elements, as in the Simpsonian blends peachza (< peach + pizza) and forfty (< forty + fifty), but this is not considered a case of overlap sensu stricto (see Renner [2019: 33] for a discussion). Table 2 provides the breakdown of the different lengths of medial phonemic overlap in the Simpsons dataset. Table 2: Distribution of English blends according to the length of medial phonemic overlap Number of overlapping segments Number of nonce blends Lexis, 14 | 2019 174 0 (e.g. grabulous < great + fabulous) 88 (38%) 1 (e.g. /s/ in croissandwich < croissant + sandwich) 65 (28%) 2 (e.g. /juː/ in youbicle < you + cubicle) 55 (23.5%) 3 (e.g. /meɪl/ in femailman < female + mailman) 17 (7.5%) 4 (e.g. /raŋk/ in Prankenstein < prank + Frankenstein) 6 (2.5%) 5 (e.g. /dekst/ in poindextrose < poindexter + dextrose) 2 (< 1%) 16 Overlap has been measured to be present in almost half of all blends (44%) in the case of institutionalized items (Renner [2019: 34]). Unsurprisingly, the proportion is higher in the case of nonce blends – 62% – as segment overlap serves to maximize the recognizability of the source elements. 2.3. Phonological split points 17 The act of clipping source elements before splicing the remaining fragments into a lexical blend leads to five different possible types of phonological split point, as illustrated in (6):2 (6a) at a syllable boundary: ce.leb.ri.[ty] + fawn.ing > ce.leb.ri.fawn.ing (6b) at an onset-nucleus boundary: Frink + [ma.n]i.ac > Frin.ki.ac (6c) at a nucleus-coda boundary: de.tec.ti[ve] + pals > de.tec.ti.pals (6d) inside a complex onset: smock + [a.p]ron > smock.ron (6e) inside a complex coda: co.bal[t] + [vi.t]a.min > co.bal.a.min (Renner [2019: 35])3 18 Table 3 shows that the distribution of the five split points among nonce blends is more marked than that of institutionalized blends (Renner [2019: 35]), with two-thirds of all non-overlap blends respecting syllable boundaries. Table 3: Distribution of split points in English non-overlap blends Location of split points Institutionalized blends Nonce blends Syllable boundary 51% 65.5% (67 items) Onset-nucleus boundary 35% 30.5% (31 items) Nucleus-coda boundary 9.5% 3% (3 items) Inside a complex onset 2.5% 1% (1 item) Inside a complex coda 2% – 19 This, again, can be read as a preference for a type of splitting that preserves syllabic constituency in order to maximize the recognition of the clipped source elements. Lexis, 14 | 2019 175 2.4. Phonological headedness 20 English blends tend to have the phonological contour of at least one of their constituents, i.e. to be parisyllabic with and have the same stress pattern as one or more source elements. This is one more structural feature that enhances the formal recognizability of those source elements which do not appear in full in the blended output. The element determining some of the properties of a whole structure may be termed the head element; here, the source element which gives its phonological contour to the blend is thus seen as its phonological head. If the head source element appears word-initially in the blend, as in the nonce blends in (7), the blend is said to be phonologically left-headed; if it appears in a word-final position, as in (8), it is right- headed: (7a) abracadabra + caramba > abracaramba (7b) avatar + turd > avaturd (8a) smock + apron > smapron (8b) stress + Cinderella > Stresserella 21 If the blend has the same phonological contour as its two source elements, it is said to display ambiheadedness, as illustrated in (9): (9) tomato + tobacco > tomacco 22 The distribution of the various patterns of phonological headedness is presented in Table 4 and contrasted with that available for institutionalized blends (Renner [2019: 42]). Table 4: Distribution of English blends according to the type of phonological headedness Type of phonological headedness Institutionalized blends Nonce blends left-headedness 24% 14.5% (34 items) right-headedness 55% 64.5% (150 items) ambiheadedness 14.5% 4.5% (10 items) non-headedness 6.5% 16.5% (39 items) 23 An overwhelming majority of nonce blends are headed – 83.5% – but the proportion of phonologically non-headed blends is markedly higher in nonce blends than in institutionalized blends. This indicates that in a relatively larger number of blending operations, maximizing the size of the two fragments in the nonce blend has been preferred to replicating the contour of either source element, as illustrated in (10): (10a) Dakota + Oklahoma > Dakotalahoma (rather than e.g. °Daklahoma) (10b) Cuba + orgasm > Cubagasm (rather than e.g. °Cubasm) 24 This illustrates a case of competition between two antagonistic recognizability factors and size maximization might be claimed to dominate contour homology more Lexis, 14 | 2019 176 frequently in nonce blending because it is a stronger agent of recognizability (phonological matter seems to be considered to be more helpful than phonological structure in the correct identification of source elements). 25 The two distributions of headed blends are similar as far as the ranking of the three types is concerned, but the gaps between types are wider in nonce blends, with an even more overwhelming preference for right-headedness. One possible partial explanation for the general preference for right-headedness over left-headedness is that it favors an alignment of the stressed syllables of the two source elements word-initially (a majority of English plurisyllabic words are stressed word-initially; this is true of an overwhelming majority of disyllabic words and of a significant majority of trisyllabic words, i.e. the two largest classes of plurisyllabic words [Clopper 2002]). 26 Finally, it is of note that a special affinity between the type of lexical shortening and the type of phonological headedness appears in the nonce-blend data, as shown in Table 5. Table 5: Distribution of English nonce blends according to the type of shortening and phonological headedness Number of left- right- ambi- non- nonce blends headedness headedness headedness headedness left-hand-side inner 27 1 2 10 shortening right-hand-side inner – 111 – 18 shortening double inner shortening 1 13 8 4 haplologic blending – 22 – 7 27 Nonce blends coined through left-hand-side inner shortening are preferentially left- headed, as illustrated in (11), while those coined through right-hand-side inner shortening and haplology are massively right-headed, as illustrated in, respectively, (12) and (13): (11a) casserole + loaf > casseloaf (11b) Nostradamus + dumbass > Nostradumbass (12a) meat + catapult > meatapult (12b) black + Frankenstein > Blackenstein (13a) clam + amphitheater > Clamphitheater (13b) Kent + entertainment > Kentertainment 28 This distribution is highly significant. In the case of left- and right-hand-side inner shortening, it shows a marked preference for a balance between segmental preservation (for the shorter source element) and contour preservation (for the longer source element) which optimizes the recognition of both source elements and the compactness of the output of blending. In the case of haplologic blending, the longer source element frequently begins with a vowel in our data, as in (13) above, and the act Lexis, 14 | 2019 177 of blending is an ingenious form of consonantal prothesis which, again, optimizes recognition and compactness. 3. Rendering English blends in French 3.1. Formal considerations 3.1.1. Translational typology 29 While there is no requirement that a blend in one language be translated as a blend in another, we decided to examine whether the 217 English blends for which a translation was available were rendered as blends in French. A total of 119 corresponding French blends were identified. We adopted a formal and coiner-oriented (i.e. translator- oriented) perspective when classifying them and consequently included several outputs where one source element was English. These bilingual blends were retained even though it is uncertain whether non-bilingual audiences would be able to identify both source elements and recognize the presence of a blend in the examples in (14): (14a) clamphithéâtre < clam + amphithéâtre (14b) ribwich < rib + sandwich 30 French translators also resorted to using a number of other morphological or non- morphological processes of lexical construction. The variety and distribution of formal subtypes is presented in Table 6. All the blends are two-element units, except for myphonies, which is not analyzed further in this section because of its marginal status. Table 6: Distribution of the formal types of translation of the 217 English blends into French Formal type Formal subtype, with an illustrative example Number Percentage Three-element blend 1 < 1% Eng./Fr. myphonies < My + iPhone + phoney Copied blend Eng. fruitopia < fruit + utopia 59 27% => Fr. fruitopie < fruit + utopie Blend Semi-creative blend Eng. didgeridon’t < didgeridoo + don’t 45 21% => Fr. didgeridiot < didgeridoo + idiot Creative blend Eng. craptacular < crap + spectacular 14 6% => Fr. merdeilleux < merde ‘shit’ + merveilleux ‘marvellous’ Eng. Wheelchairnocchio < wheelchair + Pinocchio Non-blend 98 45% => Fr. Pinocchio en fauteuil roulant ‘Pinocchio in a wheelchair’ Lexis, 14 | 2019 178 31 The French blends can be divided into three main groups. The first group is made of 59 units in which both French source elements are formal analogues of the English source elements. Due to their nature as calques, we classified them as copied blends. The second group, consisting of 45 items, contains blends where one French source element is analogous to an English source element, while the other source element is not. These blends were classified as being semi-creative as a creative translation was necessary to render the non-copied source element. The final group of blends consists of 14 blends where each of the French source elements is markedly different from each of the English source elements and the blends were thus classified as creative. 32 We also identified several cases where English blends were not translated as blends in French even though the use of analogous source elements in French would seem to pose no problem, as the examples in (15) show: (15a) Eng. Homerific (< Homer + terrific) => Fr. ºHomerifique (< Homer + terrifique) (15b) Eng. parfection (< par + perfection) => Fr. ºparfection (< par + perfection) (15c) Eng. ovulicious (< ovule + delicious) => Fr. ºovulicieux (< ovule + délicieux) 33 This underlines that rendering an English blend by a French blend remained a free choice made by the translators and that a factor like the possibility of coining a highly felicitous blend – with some segmental overlap – was not necessarily a decisive factor. 3.1.2. Lexical shortening 34 Due to their nature as copies of the English blends, the copied French blends necessarily involve the same shortening patterns in both languages and, for this reason, we shall focus only on the non-copied blends. Although the type of shortening in the semi-creative blends is influenced by the fact that they share one source element with their corresponding English blend and thus are susceptible to following the same shortening pattern, as in (16a-b), this is not necessarily the case, as can be seen in (16c): (16a) Eng. fartzilla (< fart + [God]zilla) => Fr. proutzilla (< prout ‘fart’ + [God]zilla) (16b) Eng. Nerdstrom (< nerd + [Berg]strom) => Fr. beurkstrom (< beurk ‘yuck’ + [Berg]strom) (16c) Eng. hip-hopsicle (< hip-hop + [p]opsicle) => Fr. esquimhop (< esquim[au] ‘eskimo ice cream’ + [hip-h]op) 35 When considered from the point of view of the number of shortenings alone, a distributional discrepancy is visible in Table 7: only a small minority of nonce blends – 14% – displays double shortening, compared to the 43% Renner [2019: 33] noted for institutionalized items. Table 7: Distribution of French blends according to the type of shortening Type of shortening Non-copied nonce blends Institutionalized blends Single shortening 39 (66%) 50 (53%) Lexis, 14 | 2019 179 Double shortening 8 (14%) 40 (43%) Haplologic 9 (15%) 4 (4%) Sandwich 3 (5%) – 36 This dispreference for double shortening indicates a preference for maintaining at least one source element entirely intact. 3.1.3. Phonological split points 37 We only analyzed the split points of the non-copied blends, as these, by definition, contain at least one source element unrelated to an English source element. To avoid the ambiguous analyses of blends containing overlapping segments, we determined the point of phonological splitting in the source elements of the 26 semi-creative blends and 4 creative blends where there was no overlap, giving a total of 60 splits. Splitting was determined as occurring in one of the five possible locations: at syllable boundaries (17a), between onset and nucleus (17b), between nucleus and coda (17c), in a complex onset (17d) or in a complex coda (17e): (17a) ca.ca.[to.ès] ‘cockatoo’ + boy > ca.ca.boy (17b) es.qui.m[au] ‘eskimo ice cream’ + [hip.h]op > es.qui.mop (17c) Flan.der[s] + [e]x.au.cées4 ‘granted’ > Flan.der.xau.cées (17d) fan.f[re.luche] + [ba.b]iole > fanfiole (Renner [2019: 35])5 (17e) Flan.der[s] + [e]x.au.cées ‘granted’ > Flan.der.xau.cées 38 Table 8 groups the semi-creative and creative blends together as no significant differences were observed between the two categories. Table 8: Distribution of phonological split points in French non-overlapping blends Location of split point Non-copied nonce blends Institutionalized blends Syllable boundary 88% (51 items) 47% Onset-nucleus 9% (5 items) 34% Nucleus-coda 2% (1 item) 14% Inside a complex onset – 6% Inside a complex coda 2% (1 item) – 39 The distribution demonstrates that nonce blends show a greater preference for splitting to take place at syllable boundaries rather than within syllables as this facilitates the identification of source elements and thus the comprehension of the blend. This distribution also contrasts markedly with that observed by Renner [2019: 35] for institutionalized blends, with the nonce blends showing a strong dispreference for splitting between onset and nucleus. Lexis, 14 | 2019 180 3.1.4. Phonological headedness 40 In French, phonological headedness is determined by the relationship between the length of the blend and that of its source elements, length being defined here as the number of syllables. The distribution of the data presented in Table 9 shows that the non-copied nonce blends are fairly similar to the institutionalized blends examined by Renner [2019: 43]. Table 9: Distribution of French blends according to the type of phonological headedness Type of headedness Non-copied nonce blends Institutionalized blends left-headedness 20% (12 items) 18% right-headedness 37% (22 items) 25% ambiheadedness 8% (5 items) 9% non-headedness 34% (20 items) 48% 41 Given that non-headedness means that a blend has the length of neither of its source elements, the gap between institutionalized blends and nonce blends might indicate that the latter tend to increase in length in order to retain as many phonemes from their source elements as possible. To explore this hypothesis, we examined the length of the non-copied blends relative to their longer source element. Table 10: Distribution of the length of French non-copied blends relative to the length of their longer source element Relative length Non-copied nonce blends Greater than the longer source element 32% (19 items) Parisyllabic with the longer source element 64% (38 items) Lesser than the longer source element 3% (2 items) 42 The data in Table 10 show a strong dispreference for reducing the number of syllables in a blend, with the marked overall preference being the preservation of the length of the longer source element, as in (18): (18a) Eng. bagzooka < bag + bazooka => Fr. saczooka < sac ‘bag’ + bazooka (18b) Eng. Oklasoft < Oklahoma + Microsoft => Fr. oclafoutis < Oklahoma + clafoutis 43 Retaining the length of the longer source element is a factor of increased recognizability, and especially so in the case of French, which, unlike English, cannot use the stress pattern cue in the correct identification of source elements. Lexis, 14 | 2019 181 3.2. Semantic considerations 44 To conclude this analysis of the translations of the original Simpsonian blends into French, we also sought to better appreciate the role of semantic considerations in the choice of source elements in the resultant French blends. Accordingly, the copied blends were not considered here, as they did not involve the translators making any free choice in the selection of the French source elements. Additionally, the analogous elements of the semi-creative blends – such as Godzilla in (16a) – were also not considered for the same reason. The non-analogous source elements of the French semi-creative and creative blends were divided into three groups. The first group contains blends for which the corresponding English and French source elements are synonyms, like bag and sac in (18a). The second group contains blends where the source elements are cohyponyms, like popsicle and esquimau in (16c), which are both frozen dessert foods. The final group contains the remaining items, where the corresponding source elements are not remarkably related semantically, such as nerd and beurk in (16b). A total of 73 non-analogous source elements for each language were examined. They comprised both source elements from the 14 French creative blends and their English equivalents (28 pairs), as well as the non-analogous source elements from the 45 French semi-creative blends and their English equivalents (45 pairs). Of the 73 pairs of source elements, 22 displayed a relation of synonymy and a further 22 pairs displayed a relation of cohyponymy, as shown in Table 11. Table 11: Distribution of the semantic relations between English and French source elements of non-copied blends relative to the presence of overlapping segments in French English and French Synonymous Cohyponymic Not remarkably related source elements Overlapping segments 12 22 29 No overlapping segments 10 – – 45 Table 11 also indicates whether these pairs of source elements allowed segmental overlap of the French elements because of a remarkable cross-distribution of formal and semantic types. In the cases where a French blend can be made through the use of synonyms of the English source elements, there is no clear preference between choosing source elements that allow segmental overlap, as in (19a), and source elements that do not, as in (19b): (19a) Eng. Spellympics (< spelling + Olympics) => Fr. ortholympiques (< orthographe + olympiques) (19b) Eng. Merry Fishmas (< Merry Christmas + fish) => Fr. Poisseux Noël (< poisson ‘fish’ + Joyeux Noël ‘Merry Christmas’) 46 However, in cases where a French blend contains a source element which is not a synonym of the English source element, this element always allows for segmental overlap, as illustrated in (20): Lexis, 14 | 2019 182 (20a) Eng. pray-per-view (< pray + pay per view) => Fr. paradiabolique (< para + diabolique + bolique) (20b) Eng. purrplexing (< purr + perplexing) => Fr. chat roule < chat ‘cat’ + ça roule ‘cool’ 47 As a whole, the data in Table 11 underline the importance of overlapping syllables, which help with the identification of the source elements of nonce blends. As to the striking constraint of overlapping for the translated French units not made of source elements which are synonyms of the original English source elements, the explanation might be that translators either choose to primarily respect the meaning of each original source element, in which case formal felicitousness (typified by overlapping) is secondary, or to primarily favour the formal felicitousness of the output, in which case a semantically precise translation (through the use of synonyms) is not essential. Conclusion 48 The examination of 237 English Simpsonian blends demonstrated several formal differences between nonce blends and institutionalized blends, notably in terms of the preferred type of lexical shortening and the prevalence of overlapping segments and phonological headedness. These particularities of nonce blends combine to increase the recognizability of the source elements and thus enhance the understanding of the novel output. The English nonce blends were rendered as French blends in slightly over one case in two. The latter showed highly similar tendencies to the English blends, particularly in the choice of split points, shortening patterns and phonological headedness. These results indicate that, against a widely-held view among morphologists (cf. 2.1. above), blends may not constitute a homogeneous class from a formal standpoint and that additional fine-grained studies investigating the intricacies of blending in English and other languages are undoubtedly still in order. BIBLIOGRAPHY BAUER Laurie, 2012, “Blends: Core and periphery”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre (Eds.), Cross-disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter Mouton, 11-22. BELIAEVA Natalia, 2014, Unpacking Contemporary English Blends: Morphological Structure, Meaning, Processing, PhD dissertation, Victoria University of Wellington. BELIAEVA Natalia, 2019, “Blending in morphology”, in ARONOFF Mark (Ed.-in-chief), Oxford Research Encyclopedia of Linguistics, Oxford: Oxford University Press, available at https://oxfordre.com/ linguistics/abstract/10.1093/acrefore/9780199384655.001.0001/acrefore-9780199384655-e-511 CLOPPER Cindy, 2002, “Frequency of stress patterns in English: A computational analysis”, IULC Working Papers, 2 (1), Bloomington: Indiana University. Lexis, 14 | 2019 183 HOHENHAUS Peter, 2005, “Lexicalization and institutionalization”, in ŠTEKAUER Pavol & LIEBER Rochelle (Eds.), Handbook of Word-Formation, Dordrecht: Springer, 353-373. LÓPEZ-RÚA Paula, 2004, “The categorial continuum of English blends”, English Studies, 85 (1), Abingdon: Routledge, 63-76. OED = Oxford English Dictionary Online, available at https://www.oed.com RENNER Vincent, 2015, « Panorama rétro-prospectif des études amalgamatives », Neologica, 9, Paris: Classiques Garnier, 97-112. RENNER Vincent, 2019, “French and English lexical blends in contrast”, Languages in Contrast, 19 (1), Amsterdam: Benjamins, 27-47. RONNEBERGER-SIBOLD Elke, 2006, “Lexical blends: Functionally tuning the transparency of complex words”, Folia Linguistica, 40 (1-2), Berlin: Mouton de Gruyter, 155-181. NOTES 1. The original coinage of this term has likely been captured in the earliest attestation quoted in the OED (Transactions of the Electrochemical Society [1929]): “Because of the frequent use of the terms ’adsorbed ion’ and ’adsorbed atom’ we would suggest that they be abbreviated to adion and adatom”. 2. Periods indicate syllable boundaries and the relevant syllable is underlined in case of double shortening. 3. No example of splitting at this point is attested in the dataset of English Simpsonian blends. 4. Or, more accurately, in phonological terms: [ɛ]g.zo.se. 5. No example of splitting at this point is attested in the dataset of French translations. ABSTRACTS This contribution examines the conspicuous presence of lexical blends in the long-running US television show The Simpsons and consists of two parts. The first part involves the formal analysis of 237 nonce blends in the original English-language version of the show, working on the underlying hypothesis that, despite their novelty, the audience is nonetheless able to easily decipher the blends due to a number of formal choices enhancing the recognizability of their source elements. The second part then examines the translation of these blends into Hexagonal French by taking account of formal and semantic considerations influencing whether the English nonce blends are rendered as blends in French and, if so, whether the latter display the same formal tendencies as in English. It is found that Simpsonian nonce blends notably stand out in terms of preferred type of lexical shortening and prevalence of segment overlap and phonological headedness, in both English and French. These results indicate that, against a widely-held view among morphologists, blends may not constitute a homogeneous class from a formal standpoint. Lexis, 14 | 2019 184 Cet article s’intéresse aux amalgames lexicaux de la série télévisée américaine Les Simpson et est divisé en deux parties. La première partie propose une analyse formelle des 237 occasionnalismes relevés dans la version originale anglaise de la série et se fonde sur l’hypothèse que, bien que ces amalgames soient de nouveaux mots construits inconnus, ils sont compris par les téléspectateurs du fait de plusieurs choix structurels facilitant la reconnaissance des différents éléments-sources. La deuxième partie de l’article examine ensuite les choix de traduction des amalgames anglais dans la version française de la série. Elle s’intéresse plus particulièrement à différents facteurs formels et sémantiques influençant ces choix et au degré de similitude entre les amalgames de l’anglais et du français. Il est conclu que les occasionnalismes amalgamés simpsoniens se distinguent tendanciellement des amalgames institutionnalisés, notamment pour ce qui concerne la distribution des patrons d’accourcissement, les choix de chevauchement segmental entre éléments-sources et les préférences en termes de tête phonologique. Ces résultats indiquent que, contrairement à un point de vue dominant chez les morphologues, les amalgames ne semblent pas former une classe homogène du point de vue formel. INDEX Keywords: word-formation, lexical blending, nonce word, English, French Mots-clés: morphologie, amalgame lexical, mot-valise, occasionnalisme, anglais, français AUTHORS ADAM RENWICK University of Lyon (Lumière Lyon 2) [email protected] VINCENT RENNER University of Lyon (Lumière Lyon 2) [email protected] Lexis, 14 | 2019 185 To blend so as to brand: a study of trademarks and brand names Jelena Danilović Jeremić and Jelena Josijević The authors gratefully acknowledge support from the Ministry of Education, Science and Technological Development of the Republic of Serbia (project grant 178014). […] brand names are fascinating, both in form and in meaning, because they are more consciously created than any other words. [Ingrid Piller 2005] Introduction 1 We live in an era of global consumerism, with new products or services being launched on a daily basis. To make an entry into the highly competitive market, consumer products need, first and foremost, to be named in line with the two governing principles: language economy and creativity. As Room [1994: 2] aptly put it, brand names both designate and advertise; they are created to convey information, carry desirable connotations and generate favorable perceptions. An attractive brand name can contribute to the recognition and appreciation of a particular product, providing it with a competitive advantage. Hence, brand names represent potent marketing devices, in terms of their ability to motivate customers and the quality of legal protection they are entitled to [Blackett 1998: 14]. While the packaging, price, advertising campaign, promotion, and even the product itself are subject to change, brand names are not. The familiarity and consistency of usage can result in brand names becoming widely accepted by the general public, so much so that successful linguistic creations become conventionalized (e.g. popsicle, granola, frisbee) or are used metaphorically (e.g. to hoover something up). 2 Although arbitrary collections of letters are occasionally used when new artifacts are named (e.g. Tylenol), advertising experts prefer to coin new words from old ones [Stockwell & Minkova 2001: 5], e.g. Biofreeze, Re-dew, Paingone, Yum Yum. Derivation and Lexis, 14 | 2019 186 compounding, as major word-building processes, seem to be most frequently employed in brand naming practices [Praninskas 1968: 29; Panić 2004: 287]. Blending is quite productive too, as evidenced by trade names which have developed generic usage (e.g. breathalyzer, astroturf, laundromat, vegeburger) as well as recent coinages (e.g. Snapscara, MiracOILous, or Collagenesis, to name but a few from the semantic field of cosmetics). The evident popularity of blends in brand names can be attributed to the fact that they represent not only highly creative but also playful linguistic tools that emerge as a synthesis of clipping and compounding. The combination of these two morphological mechanisms enables the formation of a unique lexical unit whose semantics is linked to both source words. As Rivkin and Sutherland [2004: 51] put it, Coined names, even those that seem completely new, rely on the hearers’ unconscious understanding of the bits and pieces of language, and their ability to transfer those new meanings to the name. Smart names choose just enough of these bits to create good feeling, leaving room for people to associate the specifics of their product with the name. 3 The loss of lexical material in blends coupled with the puzzle of novelty call for greater processing effort, which is why consumers, having figured out the intended meaning of the referent, might experience satisfaction and develop a positive attitude toward the product [Lehrer 2003: 52]. 4 Even though blending is generally thought to have increased in popularity since the middle of the 20th century, due to the growing influence of the mass media and advertising [Adams 1973: 149; Cannon 1986: 737], it can be traced to product names earlier than that. According to Gitlin and Ellis [2011: 40], Cerealine and Directoyu were introduced as corn flake clones c. 1910. In addition, Pound [1914: 21] mentions Everlastic, Locomobile and Sealpackerchief in her classification of blends from roughly the same period. In the 1930s, Berrey [1939] made a record of Feminalls, Bisquick, Tweeduroy and Playjamas. We can therefore state that blending has, for over a century now, fulfilled one of the basic marketing needs – that of naming a new product. Bearing in mind the sheer number of articles of merchandise that were introduced in the 20th century, we would expect an in-depth examination to have been performed on blends in brand names, but this does not seem to be the case. Numerous linguists make a passing reference to the productivity of blending in trade names [Marchand 1969; Hladky 1971; Adams 1973; Ungerer 1991; Piller 1995; Cannon 2000; Lehrer 2007; Lieber 2009]. Few have collected a sizeable corpus of such blends though [Praninskas 1968; Bryant 1974; Thurner 1993] and none have, to our knowledge, undertaken their detailed analysis. The aforementioned collections of lexical items can nowadays, admittedly, be considered outdated. Still, they provide valuable insight into one of the most intriguing linguistic phenomena and its distinguishing features. 5 Inspired by the comparative scarcity of pertinent studies on blending [Bauer, Lieber & Plag 2015: 462], we intend to enrich the existing body of literature with a view on blends as brand names by scrutinizing the Thurner corpus [1993], identifying the underlying motivations (phonological, graphological, stylistic) of the target blends and drawing comparisons with other relevant ‘blend-a-sets’ (e.g. Praninskas [1968]; Bryant [1974]). But first, in the next two sections, let us briefly present the linguistic sources of brand names and trademarks, and examine the role of blending in brand naming. Lexis, 14 | 2019 187 1. The creation of brand names: Key considerations 6 The terms brand name, business name, corporate name, product name, proprietary name, registered name, service name and trade name are often considered synonymous [Piller 1999: 325]. The two most widely used, brand name and trade name or trademark, are commonly used in their generic sense, to refer to any name that distinguishes an article sold or a service produced by a manufacturer from those of his competitors (Merriam-Webster online). As Lippincott and Margulies [1961] remarked almost half a century ago, tens of thousands of names are registered by the US and Trademark Office each year, which renders finding new names for products increasingly difficult. This claim appears to be even more true in an era characterized by the rise of modern technology, globalization and mass consumerism. In this multi-billion dollar industry, massive commercial efforts are nowadays invested in the brand naming process, assisted by computer programs, as well as psychological, linguistic and marketing experts. 7 Due to their similarity to proper names, commercial names tend to be capitalized, and thus constitute a separate morphological category [Baldi & Dalwar 2000: 966]. If they become popular and well-accepted, brand names can make a transition into the cultural lexicon and sometimes even change their syntactic category (e.g. Hoover, n → to hoover, v; Xerox, n → to xerox, v). In other words, they contribute to the continuing expansion of the word stock. This process is, essentially, related to the syntactic role played by brand names – as a rule, they function as proper modifiers of common nouns or noun phrases [Stvan 2006: 218], e.g. Nutter Butter cereal or Moosh dog shampoo. Incorrect practice leads to brand names being used without the accompanying noun (phrase), so they can ultimately end up as cover terms for a whole class of products. 8 When it comes to legal protectability, branding consultants commonly draw a distinction between four categories of names. In order of decreasing distinctiveness, these are: (a) fanciful or invented, (b) arbitrary, (c) suggestive, and (d) descriptive names [Blackett 1998: 10; Piller 2001: 194]. Fanciful names, such as Kodak or Pogo, are created ex nihilo and they represent the strongest, but also the least frequent type of trademark, due to their inherent distinctiveness. Arbitrary names are common words whose meaning is unrelated to their use as a trademark, e.g. Virgin as a record shop name or Orange in its use as a mobile network operator. Suggestive or imaginative names indirectly describe the product or service or refer to their attributes. Imagination is required for the inference of a connection between the name and the nature of the product or service concerned, e.g. Coppertone (suntan lotion) or Dove (soap). Descriptive names, as the term implies, describe the characteristics of the goods or services, for example Stick Fast or CarbMaster. According to Danesi [2008: 61-65], blends can be considered descriptive or suggestive names because some combinations of words describe the product make-up (e.g. Frogurt < frozen + yogurt) while others contain parts suggestive of a particular concept (e.g. Snack-tastic). 9 The formation of brand names operates under strict legal and marketing constraints. A newly introduced product or service in the US has to bear a distinctive name so it could enjoy legal protection. An oft-cited example of trademark infringement that resulted in the product name being changed is that of the Cheerios cereal, originally known as Cheerioats. Strategically desirable brand name characteristics include, besides distinctiveness, memorability, prononounceability, positive connotations, Lexis, 14 | 2019 188 suggestiveness, and euphony [Robertson 1989; Stockwell & Minkova 2001]. Although it might seem that originality plays a crucial role in the creation of a successful brand name, two opposing tendencies govern the process. In accordance with Bareš’s views [1974: 183], Baldi and Dawar [2000: 966-967] explain that: [O]n the one hand because of the need for products and services to stand out in the market, there is a call for new eye-catching patterns. On the other, convergence or analogy operates in support of an effort at uniformity, since consumers must recognize the word. This is why many brand names are coined following the same pattern. 10 For instance, many brand names contain the pseudo-scientific suffix –ex, such as Kleenex, Spandex or Durex, which exudes an air of science and scholarship due to its Latin origin. It was especially popular in the 1920s [Room 1994: 197] but seems to be favored in pharmaceutical products around the world in this day and age too (e.g. Zglobex in Serbia, Placentex in Italy). 11 Along similar lines, it is worth mentioning that the English lexicon has been expanding by means of loanwords, shortenings, composites (i.e. derivatives and compounds), blends and shifts [Algeo 1980]. The formation of trade names, which is not unlike the formation of common nouns, can be observed in accordance with the aforementioned taxonomy. French words, such as Beau or Bel(le), are used for their stylish associations [Room 1994: 5]. Cafeteria, a loanword of Mexican Spanish origin, has provided a morphological pattern copied by many business enterprises, e.g. washateria, booketeria, spaghetteria [Pyles 1952: 202]. Shortening and blending, often described as the quintessential vehicles for linguistic creativity that fall under the scope of extragrammatical morphology because they fail to conform to the regular rules and models of word creation [Fandrych 2008; Ronneberger-Sibold 2010; Mattiello 2013), produce attention-grabbing names, such as Crunchy Loggs (< Kellogg’s), Fanta (< fantasy/ fantastic), Pee-Ka-Poo (< peekaboo + poo) or Volumaniac (< volume + maniac). In general language, compounds are likely to contain two elements; in brand names, recursivity is frequently exploited with the aim of differentiating a particular product on the market through detailed description (e.g. Oven roasted chicken breast strip, Hungarian Thermal Water Mineral-Rich Atomic Heat Mask) or introducing a new variety of a well-known brand (e.g. Cocoa Puffs Ice Cream Scoop, Hubba Bubba Bubble Tape Mystery Flavor). Semantic shift, based on metaphorical and metonymic meanings, has contributed to the creation of many a brand name related to automobiles, such as Barracuda or Hot Wheels [Piller 1995]. 12 Another interesting aspect of commercial names is word play. As an essential component of the language of advertising, the deliberate exploitation of phonological or graphological similarity of words aims to grab attention, entertain, convey meanings in an economical way and individualize a product amidst a host of other similar ones [Pennarola 2003: 22; Sjöblom 2016: 462]. Such an imaginative fusion of lexical items in blends as brand names, that draws on the richness and resourcefulness of the English spelling system, poses a pleasant intellectual riddle before the target audience, e.g. VitalEyes is a play on vitalize, Seeduction on seduction. For this reason, they might experience joy upon resolving the intended meaning of the pun [Djafarova 2008: 272]. Lexis, 14 | 2019 189 2. Blending and brand naming: Theory and research 13 Blending is a rather old word-formation mechanism in English, whose playful and creative nature has been observed in the works of William Shakespeare, Charles Dickens and Lewis Carroll, inter alia. In the distant past, blends or ‘portmanteaus’ were mostly formed as puns or terms of mockery [Adams 1973: 149]. Their popularity increased considerably in the 20th century due to the growing influence of the mass media and advertising. This has led to many a linguistic discussion of their structure, meaning and function but also many a disagreement vis-à-vis their definition and position within the morphological framework [Marchand 1969; Bauer 1983; Algeo 1991; Plag 2003]. As Bauer [1983: 26] notes, delimiting blends from other morphological phenomena is no easy task because “blending tends to shade off into compounding, neo-classical compounding, affixation, clipping, and […] acronyming”. 14 Leaving aside these difficulties and inconsistencies, given that the aim of this study is to focus on the characteristics of blends in brand names in Thurner’s [1993] dictionary, his views on the topic will now briefly be discussed. In the introductory part of the book, the author [Thurner: vii] quotes a dictionary entry for blend from Webster’s Third New International Dictionary, which could be taken as a defining criterion in his corpus gathering: A word composed of parts of two words, all of one word and part of another, or two entire words, characterized invariably in the latter case and frequently in the two former cases by the single occurrence of one or more sounds or letters that appear in both the component words. 15 He states that portmanteaus can be distinguished from other types of compound words on account of their blending together shared characteristics of the source words. Also, Thurner believes that blends commonly merge the initial sounds or syllables of one word with the last of another (e.g. guesstimate), can merge like-sounding words for punning effect (e.g. sham-pagne) or even incorporate an entire word within another (e.g. metrollopis). As regards their growing popularity, he maintains that the pace of social and technological change accelerated dramatically at the turn of the century, rendering the traditional method of coining new words from roots of Greek or Latin provenance insufficient. Reliance on native language material enabled the formation of new words that could easily be introduced, understood and accepted in everyday use. Thurner’s collection of blends shows that the major sources of new words, including the fields of aerospace and military technology, animal and plant hybridization, metallurgy, the advertising industry, the television and music industries, rock music and drug culture, all resort to blending. 16 As we have already mentioned, the importance of blending in product naming was recognized over a hundred years ago [Pound 1914]. It seems to have gained momentum in the 1950s, with the rise of new inventions, processes and experiences [Bryant 1974: 163]. Linguists [Leech 1966; Marchand 1969; Adams 1973; Algeo 1980] commented on the prevailing tendency of blends to appear in newspapers, magazines and TV commercials, that is the world of advertising. As Adams [1973: 159] explained, “since they break the rules of morphology, blends have an attention-catching quality which makes them appropriate for trade names and other words in advertising copy”.1 Advertisements were, therefore, primarily used for the collection of examples relating Lexis, 14 | 2019 190 to brand names or trademarks, and their subsequent linguistic analyses. In the next few paragraphs we will examine the few relevant studies and discuss their findings. 17 There is a paucity of comprehensive research on product names in English. A notable exception is Praninskas’s [1968] corpus-based book which sheds light on the structural characteristics of blends, albeit as haplological2 or clipped compounds. She discusses numerous instances of overlapping, that is, the use of one grapheme in place of two when the final sound of the first source word and the initial sound of the second source word in a compound may be represented by the same grapheme. This phenomenon, in her view an instance of phonemicization, encompasses both traditional and altered orthography (e.g. Airefiner, Hotray vs Quicold, Starkrimson) as well as graphemes with a bimorphemic reference (e.g. Flaxoap, Quixet).3 Moreover, she notes that the compounds in her corpus frequently contain hyphens which separate the two constituent free forms, either or both of which can be clipped (e.g. Aqua-Vac, Abdo-Fit). It is also worth noting that the author found examples of clippings as constituents of compounds being freed, i.e. occurring bounded on both sides by space, e.g. Musical Rama, Vani Chest. Today, we can confirm her intuition about the potential productivity of this pattern as it abounds on the market (e.g. Choco Donuts, Coco Pops, or Nutri Love, to name but a few brand names of children’s cereal). 18 Bryant’s [1974] corpus of 306 blends contains as many as 73 trade names (24%) and many more product names that, at the time when the article was written, were not legally protected. The vast majority of them are semantically related to the fields of fashion, science and technology. It is worth mentioning that the orthographic conventions are not consistent. In addition to blends being written as single words, not infrequently a hyphen is used to mark the boundary between the source words (e.g. Tropic-nit < tropical + knit, Flare-trol < flare + control, Insta-lith < instant + lithography). Moreover, although bicapitalization has relatively recently been mentioned as a distinctive feature of Internet graphology [Crystal 2006: 93] or the high-tech sphere of language usage [Akmajian et al. 2010: 36], it is quite noticeable in the corpus material, e.g. Lubri-Cushion, Perma-Gel, BritRail. Also, some blends contain parts of company names as initial or final elements, e.g. Excel-eze (< Excello + Celanese), Monro-Matic (< Monroe + automatic), Simflex (< Simmons + flexible), Kodacolor (< Kodak + color), Recordak (< record + Kodak). In line with the general propensity of the language of advertising to playfulness and unconventionality, several blends exhibit phonetically motivated non- standard spellings (e.g. Dura-Wite, Securoslax < secure + o + slacks, Excel-eze, Polykor < polypropylene + kraftcord), make use of a phonetic realization of a particular morpheme or allomorph (e.g. Sudoprin < pseudo- + aspirin, Betarest < better + rest), contain non-English words (e.g. Cremogenized < Fr. crème + homogenized) or unusual spellings (e.g. Feathaire, Escal-aire). Some of these trade names could nowadays be considered derivatives or compounds, because they contain full words, (a vowel) and what is often referred to as a combining form or a newly emerged affix, e.g. Cruise-O-Matic, Ice-O-Matic: -matic, a combining form [McArthur 1996: 217] Perma-sized, Perma-crease: perma-, a possible derivational affix [Brinton 2000: 97] Autostereo: auto-, a combining form [Saavedra 2014: 13] Duraspeed: dura-, a combining form [McArthur 1996: 217] AromaRama: -(o)rama, a combining form [Mattiello 2017: 70]. Lexis, 14 | 2019 191 19 Bearing in mind the versatility of structural patterns observed in blends as brand names, the question which imposes itself is: how commercially successful are these blends? Cannon [2000: 954] thinks that the most successful ones are those that are morphologically transparent: For example, asphalt is a usual paving material, which, when considerably constituted of glass as the material, provides the commercial material glasphalt. Its meaning is a somewhat predictable combination of the meanings of its etyma, and is morphologically clearer than those of typical Renaissance English blends like flush (< flash + gush), perhaps necessarily because people wish to know the contents of materials that they buy, as in beefish (< beef + fish). The success of a trademark may depend on such predictability […]. 20 The aforementioned examples support the view that both phonological and graphological motivation contribute to the transparency of blends. When parts of the source words share phonemes or graphemes, it is easier to associate a blend with the source words and identify its meaning [Fischer 1998; Lehrer 2007]. For this reason, we can assume that haplology plays an important role in brand naming. This hypothesis will be tested on a corpus of approximately 600 brand names and trademarks gathered in the 1990s [Thurner 1993]. The results obtained could set the stage for explorations of more recently coined trade name blends. 3. Blends in brand names: Findings and discussion 3.1. The corpus material 21 The manually gathered corpus comprises 602 blends in brand names and trademarks listed in the Portmanteau Dictionary [Thurner 1993]. To our knowledge, this is the only reference book available that offers an abundance of examples pertaining to blending as a brand naming process. Admittedly, many of these blends are no longer in everyday use because their referents have vanished from the market over the course of time. 22 We purposely included all items labeled as brand names or trademarks in the corpus material so we could draw valid conclusions regarding their prototypical features. Blend words in the dictionary are accompanied by a brief description of the product or service, and the name of the manufacturer or parent company, e.g. Theraffin brand name Paraffin wax for use in therapy, W.R. Medical Electronics Co. 23 which enabled us to draw conclusions about the origin of blends, their structure and meaning. In relation to this, we must admit that the identification of source words was not always a straightforward task (e.g. Bitinis, Lathurn). When in doubt, we compared our interpretations with Lavrova’s [2016] or searched for the products on the Internet. Printed advertisements in newspapers or magazines, which can easily be found in online archives, provided valuable assistance. 24 The items excerpted were carefully examined in terms of their graphological, phonological, stylistic and semantic properties. To obtain numerical data, we performed a statistical analysis in the SPSS program, version 18.0. The results will be summarized in the following subsections. Lexis, 14 | 2019 192 3.2. Structural aspects of blends 25 As anticipated, the blends in our corpus feature, to a striking degree (94%), a phonological and/or graphological overlap. Taking into consideration the fact that some linguists claim that there is no overlap between the source words in most blends [Bauer 2003; Enarsson 2006], this finding indicates that blends in the language of advertising differ in certain aspects, structural complexity included, from those in general language. Table 1: Frequency of blends based on the phonological/graphological similarity of the source words Type of blend Number of blends in the corpus Overlapping 565 Non-overlapping 37 Total 602 26 A closer look at the data reveals that the vast majority of blends emerged as a result of the merging of two source words that overlapped both phonologically and graphologically. The extent of the overlap ranged from a single constituent to strings larger than a syllable, e.g. Weedigger (< weed + digger), Carpentree (< carpentry + tree), Organimals (< org anic + animals), Thunderwear (< th under + underwear), Fictionary (fiction + dictionary). Less frequently, the constituents overlapped phonologically but not orthographically or vice versa, e.g. Beenut butter (< bee + peanut), Bisquick (< biscuit + quick), Campoo (< carpet + shampoo), Gloppets (< gl ove + puppets), Intellivison (intelligent + television), Tofoodles (< tofu + noodles), Pleascent (< pleasant + scent), Plastinamel (< plastic + enamel), Pantimonium (pantyhose + pandemonium), Seaquarius (< sea + aquarius). Our results thus confirm Praninskas’s [1968] observations. More importantly, they show that the dominant pattern concerns the overlap of full words (i.e. ‘telescope blends’), that is, the merging of the hind part of the first source word with the fore part of the second source word (Table 2). Following Fandrych [2008], the various overlapping patterns can be classified into the following categories in order of decreasing frequency: Table 2: Structural patterns and their frequency in overlapping blends Number of blends in the Type of blend Example corpus overlap of full words Shampooch, Winterlude 309 initial splinter + full word with overlap Monstickers, Kaleidiskettes 130 full word + final splinter with overlap Deskretary, Legtronics 77 initial + final splinter with overlap Choclair, Immencils 42 Lexis, 14 | 2019 193 insertion of one word into the other, with Abracurldabra, 5 overlap Comfitables full word + initial splinter with overlap Creamedic 1 two splinters + full word with overlap Cosmedicake 1 27 The fact that trade names typically consist of two source words that have been seamlessly merged into one at the overlap of full words (55% of the overlapping blends), with no loss of material, suggests that semantic transparency played a key role in their creation. It seems that the manufacturers purposely opted for novel blends that require little cognitive effort when seen in print, e.g. Aliencounter, Applessence, Batherapy, Discabinet, Embracelette, Fastart, Giantarts , Infantoy , Jambrosia , Liquidose , Magicube, Nectarose, Organicurl, Practicalarm, Regallure, Selectronic, Twindow , Vinylife, Youthair, Wonderods. Transparent and immediately comprehensible, these blends testify to the domineering influence of visual stimulants on consumers’ minds. 28 As can be seen in Table 2, other patterns were also recognized in the structure of overlapping blends, some more frequent than others. Although blends are often defined as generally consisting of an initial splinter of the first source word and the final splinter of the second source word [e.g. Brinton 2000; Plag 2003; Miller 2014], our results suggest that this is not the case with commercial blends. The commonest pattern combines an initial splinter followed by a full word, e.g. Apricoating (< apricot + coating), Calendial (< calendar + dial), Shampure (< shampoo + pure), Qualitone (< quality + tone), Umbrellegant (< umbrella + elegant). Half as common were instances of a full word being fused with a final splinter, e.g. Carpetriever (< carpet + retriever), Bassitar (< bass + guitar), Dogloo (< dog + igloo), Gingeraffe (< ginger + giraffe), Gymboree (< gym + jamboree), Manwich (< man + sandwich), Petzels (< pet + pretzels), Rockoustics (< rock + acoustics), Superamics (< super + ceramics), Yardener (< yard + gardener). In some blends the phonemes of both source words are fully preserved but there is a loss of graphemes, e.g. Homade (< home + made), Locktagons (< lock + octagons), Neverinkle (< never + wrinkle), Sensurround (< sense + surround), Quiclip (quick + clip). The aforementioned prototypical pattern (i.e. initial splinter + final splinter) occurred much more rarely, e.g. Alaskimo (< Alaska + Eskimo), Crantastic (< cranberry + fantastic), Cosmerica (< cosmetic + America), Executary (< executive + secretary), Leasuramics (< leasure + ceramics), Mexicatessen (< Mexican + delicatessen), Pleasoning (< pleasant + seasoning), Tofait (< tofu + parfait), Scrunge (< scrub + spunge), Umbroller (< umbrella + stroller). The combinations of a full word and an initial splinter were practically non- existent whereas other possibilities, such as the fusion of a final splinter and an initial splinter, two initial splinters or two final splinters remained unexploited [Fandrych 2008; Mattiello 2013]. Intercalative blends, with words being inserted within a discontinuous splinter, also occurred quite rarely, e.g. Abracurldabra (< abracadabra + curl), Comfitables (< comfortables + fit), Erusticator (< eradicator + rust), Glorifried (< glorified + fry), Irrezeastables (< irresistables + freeze). Finally, we recorded a single blend consisting of two splinters and a full word, Cosmedicake (< cosmetic + medical + cake). 29 Non-overlapping blends exhibit a narrower range of structural patterns (Table 3). Lexis, 14 | 2019 194 Table 3: Structural patterns and their frequency in non-overlapping blends Type of blend Example Number of blends in the corpus full word + final splinter Soygurt, Eggcessories 18 initial splinter + full word Liquifry, Identikit 11 initial splinter + final splinter Chudge, Yogonnaise 7 full word + thematic vowel + final splinter Herbacue 1 30 Evidently, in non-overlapping blends the most common pattern involved the fusion of a full word and a final splinter, e.g. Defendamins (< defend + vitamins), Eggspendables (< egg + expendables), Funbrella (< fun + umbrella), Jagwire (< Jaguar + wire), Rooflex (< roof + complex). Less frequent were the combinations of an initial splinter and a full word, e.g. Electrocities (< electronic + cities), Extenzyme (< extensive + enzyme), Lathurn (< lathering + urn), Lubath (< luxury/luxurious + bath), Penetroil (< penetrating + oil), Smorgasgrill (< smorgasboard + grill). The same goes for the merging of an initial and a final splinter, e.g. Instamatic (< instant + automatic), Porschpoiler (< Porsche + spoiler). We also noted a single occurrence of a full word followed by a thematic vowel -a- and a final splinter (Herbacue < herb + a + barbecue). 31 Compared with other relevant data sets pertaining to blends in trade names, our results demonstrate that the overlap of full words has gained popularity as a trade naming mechanism. On the other hand, hyphenated blends that were quite common in Praninskas’s [1968] and Bryant’s [1974] studies are literally absent from the corpus material. It stands to reason that in mid-20th century, when the use of hyphens and thematic vowels in compound brand names was popular, blends too were often hyphenated. With blending becoming increasingly more common as a trade naming device, the use of the hyphen might have decreased. Another, more plausible, explanation for their absence in the corpus might be of a methodological nature. Thurner [1993] did not explicitly mention hyphenated blends in the introductory part of his book so it is possible that he considered them compounds, not blends and, consequently, decided against incorporating them into the corpus material. 32 The use of company names in brand names and trademarks, which is nowadays perhaps most easily noticeable in Nestlé’s range of products (e.g. Nescafé, Nesquik, Nespresso, Nespray, Nestea, Nestomalt, Nesplus, Nestum, Nesvita), serves as a mark of their identity, quality and good will. Historically speaking, it can be traced back to the 1930s when Berrey [1939] made a record of Shellubrication. Considering that Bryant’s corpus [1974] also includes a few examples of this sort (e.g. Kodacolor, Simflex), we expected a number of similar creations to appear in our selection of trade names. However, given the size of the corpus, their number is relatively small, e.g. Angelicare (< Angelica + care), Angelicreation (< Angelica + creation), Atlashield (< Atlas + shield), Canoah (< canoe + Noah), Charmaid (< Charma + aid), Charmaternity (< Charma + maternity), Charmour (< Charma + armour), Pandorable (< Pandora + adorable), Pulstar (< Pulse + star), Timinder (< Timark + reminder). Interestingly, in almost all of these blends the company name appears in its (clipped) form in the sinistral position (the only exception being Canoah). This is possibly related to the assumption that “the beginning letters in a word are Lexis, 14 | 2019 195 more important to process than the middle or ending letters” [Kucer 2014: 116]. A company wishing to capitalize on the existing image and reputation could therefore strategically first draw attention to its name when introducing new products or services. With every new reproduction, the company name becomes more and more easily recognizable on the market. 3.3. Stylistic aspects of blends 33 The overt inclination of the English people to linguistic play, including the exploitation of phonological similarity between words, has been documented for a long time, going back to the times of the Renaissance when the blend niniversity, echoing university, was created [Adams 1973: 148]. Brand names, as highly artificial forms, designed for commercial purposes, bring to the fore the creativity of the human mind. Fifty-six blends in the corpus make evident the freshness, originality and attractiveness of jokey names. 34 Some of them make use of homophones, e.g. Amplifire (< amplifier + fire), Charaids (< charade + aides), Delicaseas (< delicacy + seas), Dyenamite (< dye + dynamite), Eggsact (< egg + exact), Fungiside (< fungicide + side), Scentiment (< scent + sentiment), Seequence (< see + sequence). Several blends make use of the fact that the word ‘eyes’ and the verbal suffix -ize sound alike, e.g. Moistureyes (< moisturize + eyes), Scrutineyes (< scrutinize + eyes), Tantaleyes (< tantalize + eyes). 35 Many more craftfully merge similar-sounding words, e.g. Aristocat (< aristocrat + cat), Catviar (< cat + caviar), Cointainer (< coin + container), Dishtergent (< dish + detergent), Dustroyer (< dust + destroyer), Eggspert (< egg + expert), Hortisculptures (< horticulture + sculptures), Fintastic (< fin + fantastic), Nosquito (< no + mosquito)4, Parrotdise (< parrot + paradise), Pupperoni (< pup + pepperoni), Ratstaurant (< rat + restaurant), Roomance (< room + romance), Scentsation (< scent + sensation), Skintillating (< skin + scintillating), Slimderella (< slim + Cinderella). The similarity between the suffix -ity and the word ‘tea’ has been felicitously exploited in Immunitea, Maternitea, Puritea and Serenitea. 36 Compared to previous research on commercial blends, we found no instances of unorthodox spellings (for example, the use of the grapheme 3.4. Semantic aspects of blends 37 As mentioned before, blends have been identified in various semantic fields, from arts and literature to politics and government, from business and finance to science and technology. Blends in trade names have not been subjected to similar classifications and quantifications. We only have rudimentary knowledge of the popularity of blending as a naming device of food items [Danilović Jeremić & Josijević forth.; Danilović Jeremić & Josijević 2019] and pharmaceutical products [Ungerer 1991: 143]. Taking this into account, we attempted to categorize our data (Table 4) in line with the International trademark classes of goods and services [Rivkin & Sutherland 2004: 225] Lexis, 14 | 2019 196 which can be regarded as a more detailed list of semantic fields mentioned in the relevant literature on blends [Bryant 1974; Cannon 1986].5 Table 4: Most frequently occurring (semantic) classes of blends Number of blends Class Example in the corpus meats and processed foods, staple foods, natural agricultural products, light beverages, wines and Giantarts, Flavorama 68 spirits Communicenter, electrical and scientific apparatus 65 Insectocutor toys and sporting goods Crawligator, Lightarget 56 paper goods and printed matter Clipad, Grinvitations 50 cosmetics and cleaning preparations Fabulash, Organicolor 49 houseware and glass Guestray, Mercandescent 47 clothing Califamous, Footrue 38 38 Our analysis shows that blends occur frequently in the names of products that play an important role in our everyday lives: foods and beverages, electronic gadgets and devices, toys and sporting equipment, stationery, cosmetics, houseware and clothes (including footwear). These groupings accommodate 62% of all items. Other classes worth mentioning are metal goods (28 blends), furniture (28 blends), pharmaceuticals (26 blends) and paints (23 blends). This is (partially) in line with Bryant’s [1974] results as she, too, mentions fashion, home, sports and entertainment as well as science and technology as the most productive semantic fields. On the other hand, we found few examples of blends in the names of lubricants and fuels (2 blends), floor coverings (3 blends), musical instruments (8 blends) and non-metallic building materials (7 blends) while some classes of goods, such as firearms or yarns and threads, contained virtually no blends. 39 Obviously, certain classes of goods exploit blending as a naming strategy more often than others. The rapid rate of new product introductions might be forcing the companies of goods that are in high demand to exploit word coinages in an effort to catch attention and maintain their competitive position on the market [Weston & Chiu 1996]. Also, blends offer a unique possibility of merging two meanings so “they are likely to be names of ‘mixtures’ or hybrids” [Adams 1973: 158]. Eleven blends in the corpus support this claim (e.g. Accortina < accordion + concertina; Calculighter < calculator + cigarette lighter; Carboloy < carbon + alloy; Cemestos < cement + asbestos; Clamato < clam + tomato; Combrush < comb + brush; Koalaby < koala + wallaby; Raisinuts < raisin + nuts; Spoodle < spoon + ladle; Supremium < supreme + premium; Woolyester < wool + polyester). Although some companies might simply have been following trends in naming practices, we believe socio-economic aspects have been instrumental in Lexis, 14 | 2019 197 popularizing blending in brand names and trademarks. The food industry has been expanding and diversifying over the years, bringing new flavors, variants (e.g. pre- cooked, sugar-free, low-fat) or combinations of foods and beverages onto the market (e.g. Cinnamint < cinnamon + mint; Pumpernibbles < pumpernickel + nibbles; Vitamilk < vitamin + milk) because “consumers demand variety” [Weston & Chiu 1996: 22].6 Moreover, technological developments have revolutionized our way of living and working, making us dependent on a vast array of once unimaginable electronic devices and equipment (e.g. Casseiver < cassette + receiver; Fathometer < fathom + meter; Identikit < identity + kit). With more discretionary income came more possibilities for leisure and recreation time [Lazer 1994], such as numerous games, toys and sporting goods (e.g. Yardarts < yard + darts; Mechanimals < mechanic + animals; Spinsect < spinning + insect). Office supplies have increased in number and variety too, offering tools that help us manage and organize work more efficiently (e.g. Calmanac < calendar + almanac; Magicrayon < magic + crayon; Superase < super + erase). 7 So have cosmetic products which are continuously evolving, testing ingredients and launching trends in hair styles or make up (e.g. Delicare < delicate + care; Hairobics < hair + aerobics; Instantan < instant + tan). The wealth of new products in all of the aforementioned classes needed attractive, memorable and evocative names. Blending met all of these needs in a distinctive way, offering the seldom exploited possibility of merging full words to trade name creators. Apart from serving as attention-grabbers, we maintain that blends in commercial names, as newly coined words, reflected the novelty and modernity of their referents.8 Conclusion 40 Brand names are fascinating linguistic phenomena. Carefully crafted, they represent the initial step in establishing communication between companies and consumers whenever a new product or service is being launched. From a sociological standpoint, they reflect our needs, lifestyles and attitudes, shaped by the ever-changing landscape of technological advances. From a linguistic standpoint, they mirror language development: depending on the times and trends, certain patterns flourish because an innovative formation is gaining popularity (e.g. names ending in -ex, -o or -a; see Room [1994]). 41 The results of our analyses suggest that haplology plays a vital role in the creation of blends in brand names. While this is not surprising, given that similar observations regarding its function in the identification of source words have been put forward by other researchers [e.g. Fischer 1998; Cannon 2000; Lehrer 2007], we did not expect to find such an impressive number of blends involving the overlap of full words. These findings indicate that a very popular way of forming blends in the 1990s included a seamless fusion of two source words which rendered them semantically fully transparent, at least when seen in print. We can therefore conclude that visual motivation was very strong at the time – blends were to be seen as labels on the packaging or in advertising material. Also noteworthy is their playful character as a number of blends in our corpus are based on close similarity between source words or their fragments. Last but not least, as opposed to Ungerer’s [1991] point of view, we have shown that blends are systematically used not only in the names of Lexis, 14 | 2019 198 pharmaceutical products, but across a whole range of goods which form an essential part of our everyday lives. 42 Almost sixty years ago, Lippincott and Margulies (cited in Praninskas [1968: 11]) commented on “a mounting crisis in marketing – the problem of finding new names for new products”. Nowadays, with well over a million and a half active trademark registrations in the US [Johnson 2012], the situation seems to be even more challenging. Beebe and Fromer [2018] have recently put forward an empirical claim that the supply of effective trademarks is not inexhaustible because 62% of the 10,000 most frequently used words in American English have already been claimed as single word-marks. Accordingly, we can expect marketing and advertising experts to search for new naming strategies, possibly increasing the number of blends in brand names and trademarks. With our lifestyles and living conditions changing by the day, new issues are emerging (e.g. time poverty, obesity, global warming) and new commercial possibilities are arising. Some structural patterns in newly formed blends will probably become productive, leading to the emergence of new splinters and potential affixes. Furthermore, combinations of three (or more) source words, so far rather rare, could be expected to grow in number. Finally, with the rise of international markets and online trade opportunities, company names could proliferate in blends to benefit from their well-established reputation (as appears to be the case in Romanian advertising [Popescu 2015]). Further explorations of brand names and trademarks should test these hypotheses and enrich our existing body of knowledge on the dynamics of brand name change. BIBLIOGRAPHY ADAMS Valerie, 1973, An Introduction to Modern English Word-Formation, Harlow: Longman. AKMAJIAN Adrian et al., 2010 [1980], Linguistics: An Introduction to Language and Communication, Cambridge, Massachusetts: MIT Press. ALGEO John, 1980, “Where do all the new words come from?”, American Speech, Vol. 55, N°4, 264-277. ALGEO John (Ed.), 1991, Fifty Years Among the New Words: A Dictionary of Neologisms, Cambridge: Cambridge University Press. BALDI Philip & DAWAR Chantal, 2000, “Creative processes”, in BOOIJ Geert (Ed.), Morphology: An International Handbook on Inflection and Word-Formation Vol. 1, Berlin/New York: de Gruyter, 963-972. BAREŠ Karel, 1974, “Unconventional word-forming patterns in present-day English”, Philologica Pragensia, Vol. 17, 173-186. BAUER Laurie, 1983, English Word-formation, Cambridge: Cambridge University Press. BAUER Laurie, 2003 [1988], Introducing Linguistic Morphology, Cambridge: Cambridge University Press. Lexis, 14 | 2019 199 BAUER Laurie, LIEBER Rochelle & PLAG Ingo, 2013, Oxford Reference Guide to English Morphology, Oxford: Oxford University Press. BEEBE Barton & FROMER Jeanne, 2018, “Are We Running out of Trademarks? An Empirical Study of Trademark Depletion and Congestion”, Harvard Law Review, Vol. 131, N°4, 948-1043. BERREY Lester, 1939, “Newly-wedded words”, American Speech, Vol. 14, N°1, 3-10. BLACKETT Tom, 1998, Trademarks, London: MacMillan Press LTD. BÖHMEROVÁ Ada, 2010, Blending as Lexical Amalgamation and Its Onomatological and Lexicographical Status in English and in Slovak, Bratislava: ŠEVT. BRINTON Laurel, 2000, The Structure of Modern English: A Linguistic Introduction, Amsterdam/ Philadelphia: John Benjamins. BRYANT Margaret, 1974, “Blends are increasing”, American Speech, Vol. 49, N°3-4, 163-184. CANNON Garland, 1986, “Blends in English word formation”, Linguistics, Vol. 24, 725-753. CANNON Garland, 2000, “Blending”, in BOOIJ Geert (ed.), Morphology: An International Handbook on Inflection and Word-Formation Vol. 1, Berlin/New York: de Gruyter, 952-956. COOK Vivian, 2005 [2004], Accomodating Brocolli in the Cemetary, New York: Touchstone. CRYSTAL David, 2006 [2001], Language and the Internet, Cambridge: Cambridge University Press. DANESI Marcel, 2008, Why It Sells: Decoding the Meanings of Brand Names, Logos, Ads, and Other Marketing and Advertising Ploys, Lanham: Rowman & Littlefield Publishers, Inc. DANILOVIĆ JEREMIĆ Jelena & JOSIJEVIĆ Jelena, forthcoming, “Creative morphology in brand names of baby products: An analysis of compounds and blends”. DANILOVIĆ JEREMIĆ Jelena & JOSIJEVIĆ Jelena, 2019, “A cerealicious brekki: A linguistic analysis of children’s cereal boxes”, Paper presented at the international conference BELLS 90, May 30-June 1, Belgrade: Faculty of Philology. DJAFAROVA Elmira, 2008, “Why Do Advertisers Use Puns? A Linguistic Perspective”, Journal of Advertising Research, Vol. 48, N° 2, 267-275. ENARSSON Anna, 2006, New Blends in the English Language, Karlstad: Karlstad Universitet. FANDRYCH Ingrid, 2008, “Submorphemic elements in the formation of acronyms, blends and clippings”, Lexis, Vol. 2, 103-121. FISCHER Roswitha, 1998, Lexical Change in Present-Day English: A Corpus-Based Study of the Motivation, Institutionalization, and Productivity of Creative Neologisms, Tübingen: Gunter Narr Verlag. GITLIN Martin & ELLIS Topher, 2011, The Great American Cereal Book: How Breakfast Got Its Crunch, New York: Abrams Image. HLADKY Josef, 1971, “The orthography of British trade names”, Sborník prací filozofické fakulty brnĕnské university, A19, 145-160. JOHNSON Christopher, 2012, Microstyle: The art of writing little, London/New York: W. W. Norton & Company. KUCER Stephen, 2014, A Conceptual Base for Teaching Reading and Writing in School Settings, New York/ Abingdon: Routledge. Lexis, 14 | 2019 200 LAVROVA Natalie, 2016 [2009], Англо-русский словарь Современные тенденции в словообразовании: Контаминанты, Mосква: Флинта. LAZER William, 1994 [1987], Handbook of Demographics for Marketing and Advertising: New Trends in the American Marketplace, New York: Lexington Books. LEECH Geoffrey, 1966, English in Advertising: A Linguistic Study of Advertising in Great Britain, London: Longmans. LEHRER Adrienne, 2003, “Understanding trendy neologisms”, Rivista di Linguistica, Vol. 15, N°2, 371-384. LEHRER Adrienne, 2007, “Blendalicious”, in MUNAT Judith (Ed.), Lexical Creativity: Texts and Contexts, Amsterdam/Philadelphia: John Benjamins, 115-133. LIEBER Rochelle, 2009, Introducing Morphology, Cambridge: Cambridge University Press. LIPPINCOTT Gordon & MARGULIES Walter, 1961, “How to name a product”, Management methods, Vol. 2, 47-51. MARCHAND Hans, 1969 [1960], The Categories and Types of Present-Day English Word-Formation: A Synchronic-Diachronic Approach, München: Verlag C.H. Beck. MATTIELLO Elisa, 2013, Extra-grammatical morphology in English: Abbreviations, Blends, Reduplicatives, and Related Phenomena, Berlin/Boston: Walter de Gruyter. MATTIELLO Elisa, 2017, Analogy in Word-formation: A Study of English Neologisms and Occasionalisms, Berlin/Boston: Walter de Gruyter. MCARTHUR Tom, 1996, The Concise Oxford Companion to the English Language, Oxford: Oxford University Press. MILLER Gary, 2014, English Lexicogenesis, Oxford: Oxford University Press. PANIĆ Olga, 2004, “Brand names: How they are made and what they are made for”, British and American Studies Journal, Vol. 10, 285-291. PENNAROLA Cristina, 2003, Nonsense in advertising: Deviascion in English Print Ads, Napoli: Liguori Editore. PILLER Ingrid, 1995, American Automobile Names, PhD thesis, Dresden: Technische Universität, also available at http://www.languageonthemove.com/wp-content/uploads/2011/02/ piller_1995_american-automobile-names.pdf PILLER Ingrid, 1999, “Iconicity in brand names”, in NANNY Max & FISCHER Olga (Eds.), Form miming meaning: Iconicity in language and literature, Amsterdam: John Benjamins, 325-341. PILLER Ingrid, 2001, “Brand Name Formation”, Nomina Africana, Vol. 15, N°1-2, 189-202. PILLER Ingrid, 2005, “Naming”, in STRAZNY Philip (Ed.), Encyclopedia of Linguistics, New York: Fitzroy Dearborn, 724-726. POUND Louise, 1914, Blends: Their Relation to English Word Formation, Heidelberg: Carl Winter’s Universitätsbuchhandlung. PLAG Ingo, 2003, Word-formation in English, Cambridge: Cambridge University Press. POPESCU Alina Raluca, 2015, “On the portmanteau words based on commercial names in current Romanian advertising”, Diversité et Identité Culturelle en Europe, Vol. 12, N°2, 215-232. PRANINSKAS Jean, 1968, Trade name creation: Processes and patterns, The Hague: Mouton. Lexis, 14 | 2019 201 PYLES Thomas, 1952, Words and Ways of American English, New York: Random House. RIVKIN Steve & SUTHERLAND Fraser, 2004, The making of a name: The inside story of the brands we buy, New York: Oxford University Press. ROBERTSON Kim, 1989, “Strategically desirable brand name characteristics”, The Journal of Consumer Marketing, Vol. 6, No. 4, 61-71. ROOM Adrian, 1994 [1991], NTC’s Dictionary of Trade Name Origins, Lincolnwood: NITC Business Books. RONNEBERGER-SIBOLD Elke, 2010, “Word creation: Definition – Function – Typology”, in RAINER Franz et al. (Eds.), Variation and Change in Morphology: Selected Papers from the 13th International Morphology Meeting, Vienna, February 2008, Amsterdam: John Benjamins, 201-216. THURNER Dick, 1993, Portmanteau Dictionary: Blend Words in the English Language, Including Trademarks and Brand Names, Jefferson: McFarland & Company, Inc. SAAVEDRA CORREIA David, 2014, “Automatically identifying blend splinters that are morpheme candidates”, Digital Scholarship in the Humanities, Vol. 31, N°1, 55-71. SJÖBLOM Paula, “Commercial Names”, in HOUGH Carole (Ed.), The Oxford Handbook of Names and Naming, Oxford: Oxford Universiy Press, 453-464. STOCKWELL Robert & MINKOVA Donka, 2001, English Words: History and Structure, Cambridge: Cambridge University Press. STVAN Laurel, 2006, “The contingent meaning of –ex brand names in English”, Corpora, Vol. 1, N°2, 217-250. UNGERER Friedrich, 1991, “Acronyms, trade names and motivation”, AAA: Arbeiten aus Anglistik und Amerikanistik, Vol. 16, N°2, 131-158. WESTON Fred & CHIU Susan, 1996, “Growth Strategies in the Food Industry”, Business Economics, Vol. 31, N°1, 21-27. NOTES 1. The Learn to speak Snacklish ad campaign, which, for the most part, made use of lexical blends in the distinctive Snickers colors on a chocolate brown background, proves this point (e.g. snaxi, chewniversity, satisflying, chocollege, chewpiter, yumazing, social nutwork, chompensation, chompion, feedquipment, satisfectellent). 2. Adams [1973: 150] defines haplology as the overlapping of vowels, consonants or syllables, e.g. privilegentsia (< privilege + intelligentsia) or selectorate (< select + electorate). 3. We capitalized examples, originally presented in small caps, from Praninskas [1968] to achieve uniformity throughout the paper. 4. Interestingly, the only occurrence of the negative determiner in English blends was recorded in Magnox [Renner cited in Böhmerová 2010: 81]. 5. Blends occurred in the names of few services (e.g. Bestemps, an employment service; Domesticare, a home-cleaning franchise; Frostop, a fast-food franchise; Travelodge, a hotel/motel chain) so we decided to exclude them from the results presented in this part of the paper. 6. According to recent estimates, the food manufacturing industry in the US generates a revenue of approximately $760 billion per year (see https://www.cmtc.com/blog/food-and-beverage- manufacturing-trends-and-challenges-2016). Lexis, 14 | 2019 202 7. For more information about the development of stationery stores see https:// www.referenceforbusiness.com/industries/Retail-Trade/Stationery-Stores.html 8. A more detailed semantic investigation focusing on the relationships between the constituents of the target blends falls outside the scope of this paper due to the size of the corpus. ABSTRACTS Brand names represent valuable linguistic assets. They serve a variety of purposes, from product differentiation to corporate identity. Creativity and wit are employed in brand naming practices, frequently resulting in the formation of blends. Although blending has long been recognized as a prominent feature of advertising and marketing discourse, next to no research has hitherto been devoted to blends in brand names. Hence, the aim of this paper is to analyze approximately 600 brand names excerpted from the Portmanteau Dictionary [Thurner 1993]. Having examined the phonological, graphological, stylistic and semantic motivations, we conclude that the distinguishing features of blends in brand names are the overlap of full words and word play. In terms of particular classes of goods, blends seem to permeate the names of foods and beverages, electrical and scientific devices, toys and sporting equipment, as well as stationery, cosmetics and houseware. Les noms de marque représentent des atouts linguistiques indéniables et remplissent de multiples fonctions, de la différentiation du produit jusqu’à la création et/ou le renforcement de l’image de marque de la société. La créativité linguistique et les jeux de mots sont convoqués lors de la création d’un nom de marque, qui prend fréquemment la forme d’amalgames. Bien que le rôle du phénomène d’amalgamation ait été depuis longtemps reconnu comme une caractéristique dominante du discours publicitaire et du discours du marketing, quasiment aucune recherche n’a été consacrée jusqu’à présent au rôle des amalgames dans les noms de marque, d’où l’objectif de cet article, qui consiste à analyser les quelque 600 noms de marque trouvés dans le Portmanteau Dictionary [Thurner 1993]. Après avoir étudié les motivations phonologiques, graphiques, stylistiques et sémantiques, nous concluons que les traits distinctifs des amalgames dans les noms de marque résident dans le chevauchement de mots entiers et de jeux de mots. En ce qui concerne les types de produits concernés, il semble que les amalgames s’infiltrent aussi bien dans les noms de boissons et nourritures, d’appareils électriques et scientifiques, de jouets et de matériels sportifs que dans les noms de cosmétiques et d’articles ménagers. INDEX Mots-clés: amalgamation, noms de marques, marques déposées, corpus Thurner Keywords: blending, brand names, trademarks, the Thurner corpus Lexis, 14 | 2019 203 AUTHORS JELENA DANILOVIĆ JEREMIĆ University of Kragujevac [email protected] JELENA JOSIJEVIĆ University of Kragujevac [email protected] Lexis, 14 | 2019 204 Gold Punning: studying multistable meaning structures using a systematically collected set of lexical blends Daniel Kjellander Introduction and aims of the study 1 It has been claimed for almost a century that lexical blending is an increasing language phenomenon, especially in the domain of English [Pound 1914; Bryant 1974; Kemmer 2003; López Rúa 2010]. At the same time, several studies report its uncertain, or even marginal, status in linguistic research [Cannon 1986; Kemmer 2003; Brdar-Szabó & Brdar 2008; Fandrych 2008]. Lexical blending would therefore be expected to be an attractive domain of word formation research, as it contains numerous possibilities to explore new ground in this particular field of linguistics. Nonetheless, the well attested complexity of blends, phrased already more than a century ago in Bergström [1906: 27, § 49] as “[h]ere the capriciousness of language operates with great freedom”, seems to have remained an obstacle in efforts to address lexical blending for the larger part of the 20th century (see for instance Bauer [1983]; Cannon [1986]). 2 It would be no exaggeration to say that the depiction above constitutes a challenge to contemporary word formation research [Cannon 1986; Dressler 2000; Bauer 2012; Gries 2012], and the field has in fact attracted a fair amount of interest in recent years. The increasing number of studies in the last three decades bear witness to this, and the broadened theoretical scope, together with the development of novel tools for data analysis, further emphasize the dynamic character of this field of research. 3 Despite numerous efforts, the patterns, the distribution, and the motivations of blends have remained unclear in much of previous research [Gries 2004b, 2006]. The difficulty to collect blends in a systematically stringent manner is likely one reason behind this, and as the title indicates, gold panning is suggested as a metaphor for the hard and Lexis, 14 | 2019 205 time-consuming work to extract lexical blends from a large body of textual data. Importantly, punning, as in Gold Punning, is understood here broadly as a strategy to achieve effects of wordplay which is an intrinsic characteristic of lexical blending [Lehrer 1996; Gries 2006; Alm-Arvius 2012; Renner 2015]. Instantiations range therefore from potentially humorous expressions such as relationsshit (relationship + shit) to more commonplace items such as glocal (global + local). 4 Although there are problematic aspects of blend research, the existing literature addresses constructively a number of topics concerning the nature of blends, and some of these have resulted in promising steps forward towards more comprehensive, and more empirically grounded, theories on lexical blending. Among assumptions that have become more established in terms of general agreement are the tendencies for source words to display structural (graphemic and phonemic) similarity [Kelly 1998; Gries 2004a; Fábregas & Scalise 2012], the categorical difference between blends and complex clipping [Bat-El 2006; Gries 2012; Arndt-Lappe & Plag 2013; Beliaeva 2014], and the distinction between speech error blends arising from conflicting lexical activation and word formation blends that are the result of intentional lexical processes1 [Gries 2012]. 5 Other issues seem to require yet more investigation. For instance, there is no general agreement as to the categorical and terminological relation between blends and compounds. Fábregas and Scalise [2012] argue in favor of a categorical distinction based on the unanalyzability of blend segments and the recurring phenomena of non- linearity and structural overlap of source words in lexical blends. This is contrasted with accounts assuming straightforwardly that blends constitute a special case of compounding [Ronneberger-Sibold 2012; Arndt-Lappe & Plag 2013]. Another question concerns dynamic processes in which lexical blends are the source of further blended forms, and sometimes even produce segments that acquire morphematic qualities [Soudek 1971; Warren 1990; Lehrer 1996, 1998, 2007; Beliaeva 2014]. Viewpoints range from attitudes assuming that seriality in any form is incompatible with the concept of lexical blending [Fradin 2015] to approaches including most items that display some sort of truncation and fusion of form [Plag 2003; Lehrer 2007]. To address such complex issues of categorization, it has been suggested that a prototype theory perspective is appropriate to account for the complexity of lexical blending [Brdar-Szabó & Brdar 2008; Renner 2015]. Because prototype theory is commonly applied in a Cognitive Linguistics (CL) framework [Taylor 2003; Geeraerts 2006c; Ungerer & Schmid 2006], it seems natural that several accounts have put forth CL as an ideal theoretical outset when studying blends [Kemmer 2003; Gries 2006; Silaški & Đurović 2013]. 6 Further examples of issues that remain largely unresolved are the presumed semantic relatedness between the lexemes that produce the constituent segments, traditionally called the source words (SW), of a blend [Algeo 1977; Brdar-Szabó & Brdar 2008] and the prototypicality of certain formation patterns such as distribution of truncation strategies and source word overlap [Renner et al. 2012; Arndt-Lappe & Plag 2013]. 7 Up until the 1990s, most accounts were concerned with classification and categorization of lexical blends [Kemmer 2003], and contemporary work has thus shifted the focus of attention towards the structure of blends, and efforts have been made to provide statistical evidence to arrive at more empirically robust conclusions [Gries 2004b, 2006]. On good grounds, it is thus reasonable to say that the knowledge about lexical blends has increased substantially in the last thirty years. It is, however, also apparent that many questions await further and more elaborate answers, which is Lexis, 14 | 2019 206 especially true as to the semantics of blends and the motivations for their existence [Gries 2006; Bauer 2012; Gries 2012]. 8 A recurring issue related to a large part of previous studies is how to collect data in a systematic fashion so as to ensure that the findings and conclusions are, if not generally applicable, at least allow replication and representativity within stipulated limitations. As will be discussed below, the present study departs from the standpoint that this is indeed a crucial concern in contemporary blend research. 9 The aim of the current study is two-fold. First, a methodological approach is suggested that enables quantitative and qualitative analyses based on a systematically collected dataset. The importance of developing systematic data retrieval procedures in blend research can hardly be overestimated, which is why this aim has been given slightly more prominence in this account. Central issues concern the representativity of the data, their relevance and potential for generalizability, and delimitations related to pragmatic aspects. For instance, if the data originate from a specific jargon, it cannot be taken for granted that usage patterns are automatically transferable to other domains of language. Therefore, the source of the data, involving aspects such as temporal scope, genre, mode, and culture, is crucial. 10 Second, there are several reports on the influence from general cognitive mechanisms on the formation and use of lexical blends [Kelly 1998; Kemmer 2003; Gries 2004b; Kjellander 2018]. Recent findings show a number of observable structural tendencies in blends [Gries 2012; Arndt-Lappe & Plag 2013; Beliaeva 2014; Juhasz et al. 2016], but indications of influence from more general cognitive processes also suggest that a wider scope is necessary to understand lexical blends better [Kelly 1998; Kemmer 2003; Gries 2004b; Fandrych 2008; Bauer 2012; Ronneberger-Sibold 2012]. The second aim of the study is therefore to investigate a particular semantic aspect of the blends in the data, namely how lexical ambiguity is exploited as a means to achieve certain communicative effects. A central research question that was applied for this purpose is phrased: how is lexical ambiguity realized in lexical blends in a contemporary US web news context? 11 Ambiguity in this context is understood as lexical structures allowing, or even driving, meaning construal involving unresolved semantic conflicts. While ambiguity is a language phenomenon sometimes evoking interest and humorous reactions, but perhaps more commonly thought of as a communicative problem, it is investigated what potential purposes it might serve in lexical blending. The theoretical outset, a Cognitive Linguistics (CL) perspective, means that meaning construal is seen as a basis for all semantic operations concerned with the conceptualization of utterances and comprehension [Langacker 2008]. Meaning construal is furthermore understood as a dynamic process, in which multiple modalities come into play [Geeraerts 2006b], and ambiguity influencing meaning construal can thus be described as an inherently cognitive mechanism drawing on general cognitive abilities. In short, this mechanism is understood as a cognitive constraint influencing the formation and use of lexical blends (cf. Gries [2004b]; Kjellander [2018]). 12 The aims of the investigation are pursued within the context of US web news between January 2010 and March 2018. Because the intention of the study is to study patterns of present-day blend usage, a large contemporary database is required, which is why the entire corpus up to the time of the data download was chosen. This means that 99 months of web news (or approximately 6 billion words) are available, and, Lexis, 14 | 2019 207 subsequently, that the requirements as to quantity and focus on contemporary language use are met. 13 For reasons of space, only common nouns as potential sources of lexical blends are investigated in this study. Thus, five limitations apply to the material and the findings of the study: i) the empirical material is restricted to US data, ii) the genre is news texts published on the Internet, iii) the mode is written language in an online setting, iv) the temporal scope applies as above, and v) the results apply to patterns of blend formation from common nouns as they are defined in the present study. 1. Definitions 14 Many attempts have been made to define lexical blends, but as matters stand only preliminary attempts are at hand [Bauer 2012; Renner et al. 2012; Beliaeva 2014], and Kemmer [2003: 71] argues that “[l]exical blends are so varied in form that no neat taxonomy can do justice to the full range of the phenomenon.” Nonetheless, a very brief but relatively uncontroversial description of a lexical blend can be summed up as follows: 15 A lexical blend is a lexeme • formed from two (or rarely several) source words, of which either one or both are truncated in the blended lexical construct. If more than one source word is truncated, the final part of the first source word, and the initial part of the second source word are merged. • partly motivated by the fusion of structure as well as the fusion of conceptual content of its source words, which sometimes includes shared structure, be it graphemically, phonemically, or both. 16 There are, however, a number of problems with compressed definitions of this type. First, truncation patterns are reportedly diverse, and there is an intriguing interplay between truncation and the sharing of graphemes and phonemes [Kemmer 2003; Lalić- Krstin & Silaški 2018]. For instance, the structure of a blend such as dragula (drag + dracula) involves an uncertainty as to the origin of the blend. It is, in principle, not possible to tell which of the source words is responsible for the segment dra-, and the truncation pattern can therefore not be determined, at least not as a result of a structural analysis [Kemmer 2003]. Likewise, the blend robocup (robocop + cup) is ambiguous as regards the segment -cup, and this blend is made further complex by the fact that the first source word is in itself an amalgamated form derived from robot and cop. 17 Second, the status of the segments differs as well. Transparency typically varies, and it is common for segments to be reused in forms derived from already existing blends [Kemmer 2003; Lepic 2016; Lalić-Krstin & Silaški 2018]. As indicated above, the recycling of already existing blends, or segments of blends, is a field requiring further research, and the workings of the mechanisms behind these processes are still in need of more detailed analyses. 18 Third, the meaning potential of a lexical blend is inherently difficult to account for. Although there are observations of iconic qualities in blends such as smog [Ronneberger-Sibold 2006] and priming effects in the process of decoding [Lehrer 1996; Gries 2006], the semantics of lexical blends remains largely unexplored. For instance, only specific aspects of the source words of dragula and robocup are intended to fuse so Lexis, 14 | 2019 208 as to result in an intelligible blend. This means that the activation of a certain conceptual subdomain, thus metonymically motivated, is often necessary for successful decoding. The title of the present text obviously draws on this principle; evoking the domain of gold panning may be interpreted as an intention to assign a high value to the object of investigation. This is, however, not what it is supposed to convey (although the reader’s interpretation is of course not limited to any particular construal of its meaning). As will become apparent in the following sections, the intended meaning is instead oriented towards the demanding process of retrieval, and the selection of a specific aspect of the source expression is used as an illustration of the metonymic qualities commonly observed in lexical blends. 19 These three aspects – diversity of truncation patterns, dynamicity of blend segments, and rich and complex meaning potential – are successfully accounted for within a Cognitive Linguistics approach, where the prototype organization and the fuzzy boundaries of linguistic categories are acknowledged and addressed [Taylor 2003; Langacker 2008]. 20 Besides the semantic, or conceptual, gap in blend research, methodological discussions concerned with the collecting of blend data are scarce in previous accounts. There are several possible reasons for this. The presumed marginality of blends in combination with i) their ephemeral character [Cannon 1986; Lehrer 1996; Brdar-Szabó & Brdar 2008; Lalić-Krstin & Silaški 2018], ii) their informality [Cannon 1986; Fandrych 2008; López Rúa 2012], and iii) their structural complexity [Bergström 1906; Bauer 1983; Cannon 1986; Bauer 2012; Renner 2015] means that they are very hard to capture in any ordered and transparent way. In addition, despite the claims that blends are increasing, they are nonetheless infrequent in comparison with other lexical processes. As Arndt-Lappe & Plag [2013] point out, previous accounts of blending have not satisfactorily taken frequency into account, which underscores the need for a more robust empirical background to generalizations concerned with the use and distribution of blends. 21 It could be argued that there is in fact a vicious circle influencing much of earlier research; blends are portrayed as irregular and unsystematic, which presumably makes them difficult to retrieve systematically. As a consequence, they appear all the more ephemeral, informal, and complex, which further underscores the seemingly unsurmountable objective of systematicity. 22 Fortunately, there are two distinctive trends in contemporary blend research that promise to push the development in new directions; theoretical and technological progress. The theoretical development of Cognitive Linguistics since the 1980s has opened up the field of linguistics for broader, more dynamic, and more inclusive analyses of language, which explains the nature of lexical blends significantly better than structuralist or generativist views on language [Kemmer 2003; Gries 2006; Fandrych 2008; Silaški & Đurović 2013]. Furthermore, the framework of CL in combination with digital corpora, powerful software, and precise psycholinguistic tools offer unprecedented opportunities to explore lexical blending in the 21st century. 2. Data and Method 23 It follows from the aims of the study that the data are not only accounted for in terms of sources and limitations, but also that the process of retrieving data is a core methodological aspect in itself. The current section is therefore divided into two Lexis, 14 | 2019 209 subsections aiming to reflect this characteristic of the study. Thus, section 2.1. contains general descriptions of methodological choices and delimitations, while section 2.2. is a procedural account describing in detail the various steps of the investigation. 24 A comment will also be made here on a few terminological topics. First, descriptions of truncation patterns in blends usually employ the letters AB and CD to represent the (typically two) source words [e.g. Beliaeva 2014]. The initial segment of the SW is indicated with A and C respectively, disregarding the length or other characteristics of the segment. The same rationale is valid for the final SW segment represented by B and D. Brunch (breakfast + lunch) is therefore an AD structure, whereas digicam (digital + camera) is described as AC. As long as this convention is applied as a schematic structural description of truncation, it serves its purpose well. The only diversion from this coding pattern is the added sign “|” in “A|BC|D”, which helps to signal the uncertain characteristics of blends where no structural material seems to be removed (e.g. stoption from stop + option). 25 Second, the topic of this investigation is concerned with lexical blends. In analogy with compounds and compounding, these items are seen as instantiations of the process of lexical blending. The term blend is sometimes used as an abbreviation of lexical blend, while other potential terms, such as conceptual blend, speech error blend, and syntactic blend, are never intended in the abbreviated form. 26 Finally, section 2.2. contains a certain amount of technical terms needed to account for the procedures of the data collection. These terms are kept at a minimum, and definitions are made as generic and brief as possible, focusing solely on their relevance for the present study. 2.1. Methodological choices and limitations 27 Examples of data collection procedures in previous studies contain descriptions such as “[t]he following pages grew […] out of some chance notations of blends made from time to time by the author” [Pound 1914: Prefatory note] and ”[m]any of the examples cited here are from these studies; others are of my own collecting” [Algeo 1977: 47]. Although there are examples of more transparent accounts of the origin of the data, especially in more recent studies, there are few investigations that present data collection procedures allowing in depth analyses of the empirical material in relation to other aspects of language. As a result, discussions on, for instance, i) the semantic relation between source words, ii) the distribution of lexical blend characteristics, iii) conclusions concerning the diachronic development of lexical blending, and iv) genre distribution remain topics that often do not go beyond assumptions in need of more robust empirical grounding in order to be developed further. To some extent, this may explain some of the gaps in blend research that have continued to be unresolved issues for decades. According to Fandrych [2008], the seemingly unsystematic nature of blends has even licensed a neglect of the phenomenon in linguistic investigation, especially in the generative tradition. On the other hand, the difficulties to collect these ephemeral, informal, creative, and complex items in systematic ways are indeed considerable, which is why the emergence of digital media and computer processing have proved to be crucial factors in attempts to establish a better understanding of lexical blending. Last but not least, the CL orientation towards language in use implies that the focal scope encompasses contextual and general cognitive factors. This means Lexis, 14 | 2019 210 that the requirements on the usage data are considerable, but also that the data are potentially informative as regards the processes involved with lexical blending. 2.1.1. Lexical blends and Internet language 28 Besides the possibility to benefit from the efficiency of digital tools, the choice to investigate data in the domain of Internet language is motivated by, at least, two linguistic characteristics of the domain itself. First, it has been observed that the traditional division between spoken and written mode does not easily apply to Internet language. Crystal [2011: 17] argues that it can be described as “writing which has been pulled some way in the direction of speech”, and in Crystal [2001] it is discussed whether Internet language should in fact be thought of as a mode in itself, separated from spoken and written representations. Second, there are numerous observations of the non-standard (sometimes even called ‘rebellious’) spelling conventions of the Internet [Shaw 2008]. Conventional views on misspelling are thus challenged by (sometimes ideologically motivated) orthographic representations that divert from traditional norms [Sebba 2003]. 29 Both these aspects, the intermediary nature of the modality and the non-standard spelling, dovetail important aspects of lexical blends. As Lalić-Krstin and Silaški [2018] demonstrate, blends sometimes display a discrepancy between their graphemic and phonemic representation in expressions such as Czechout (Czech+ checkout), and Kjellander [2018] shows how the language users may choose to foreground either phonemic or graphemic efficiency in a pair of related lexical blends rapepublican and rape-ublican. These forms are used similarly with similar meanings (derived from rape + republican), but while rapepublican favors graphemic decoding (because of the gemination it causes when pronounced, see Spencer [1996]), rape-ublican appears less transparent graphemically while pronunciation is smoother. 30 The freedom of the language user seems thus considerable in the formation and use of lexical blends [Bauer 1983; Cannon 1986]. At the same time, it has been claimed that there are constraints that influence where source words are split [Kubozono 1990; Kelly 1998; Gries 2004b] and in which linguistic domains blend tend to occur [Ronneberger- Sibold 2006, 2012]. In addition, the prosody of the source words has been shown to have consequences for the form that a blend takes [Arndt-Lappe & Plag 2013; Beliaeva 2014]. Despite such seemingly restrictive mechanisms, it would still be odd to claim that a certain blend form is incorrect in a traditional sense of the word. Instead, the freedom of the language users typically leaves room for considerable variation, while the observed structural patterns are best described as probabilistic bottom-up tendencies identifiable in their shape [Gries 2012]. The creativity and flexibility of blend formation seem thus to be particularly well aligned with similar observed tendencies in Internet language. It should therefore come as no surprise that Internet language is often referred to as a fertile ground for the formation and use of lexical blends [Lehrer 2007; Cook & Stevenson 2010]. 2.1.2. Limitations 31 Previous observations of the distributional characteristics of blends have shown that they appear more markedly in certain domains and genres, e.g. advertising, electronic media, and various types of informal contexts [Cannon 1986; Ronneberger-Sibold 2006; Lexis, 14 | 2019 211 Lehrer 2007; Fandrych 2008]. Because of this uneven distribution, the current study assumes that it cannot be taken for granted that observed patterns in lexical blends can be automatically transferred from one domain to another. Instead, it is considered important to delimit the field of investigation in this respect, with the objective to study systematically one relatively delimited language domain, and at a later occasion possibly compare the findings between genres so as to arrive at more wide-ranging conclusions. In addition, the same data collection procedure could be applied comparing different temporal limitations to explore possible diachronic patterns empirically, and the regional scope could be exchanged so as to compare the output from several parts of the English-speaking world. 32 The chosen genre limitation of the present study is English written news media, which has been claimed to harbor a certain amount of lexical blends [Lehrer 2007], and was therefore considered a good starting point. For reasons presented in the previous section, the empirical material was furthermore limited to online data. It can, however, easily be imagined that these limitations imply a quantitative level of information management that potentially exceeds any reasonable primary estimation of workload. Therefore, it was necessary to work with a systematically organized corpus including inbuilt regional and temporal limitations, and to this purpose the Brigham Young University Corpus News On the Web2 (NOW) was chosen to constitute the source of the data. 33 Finally, in order to further limit the output, it was decided to take common nouns as a starting point for the investigation. Apart from general observations of grammatical functions of collected lists of blends and their source words, the interplay between grammatical categories and lexical blends also remains a field in need of further research. Although the present investigation can say little about this aspect within its designated scope, a well delineated dataset in this respect means that it can in principle be compared with future studies aiming at research questions along these lines. 2.2. Data collection procedure 34 The data source of the study, the NOW corpus, is a digital corpus continuously collecting texts from 20 English speaking regions in the world. The online interface allows various types of restrictions and filters to enable a wide array of linguistic queries related to, for instance, temporal, regional, grammatical, and lexical limitations. Figure 1 exemplifies a simple filtered online query for the word blend used as a verb in the US region. 35 In the present study, however, the requirement for detailed and multifaceted query restrictions resulted in unstable data outputs and repeated server breakdowns, which eventually led to the decision to download the entire corpus for offline management. This would ensure the stability of the data as well as providing a sufficient level of computer processing power needed to run required queries. Lexis, 14 | 2019 212 Figure 1: The NOW corpus online user interface Image 1038DB5800009B2000005899F72D9D240F6DA205.emf A download was thus performed in April 2018. At this time, the size of the corpus was approximately 6 billion words, and the temporal scope ranged from January 2010 to March 2018, which meant that 99 months of collected online data was included in the empirical material. The corpus came in compressed folders containing text files tagged for region, year, and month (e.g. “15-02-us.txt” for February 2015, US data), which made limitations of these variables easy to manage. Only the data from the United States were retained for the purposes of the current investigation, which resulted in a size reduction to slightly more than one billion words. The NOW data were then imported into SQL-tables3 in the PostgrSQL 4 management software PGadmin 5. These operations were necessary as the offline version of the corpus did not come with a user interface such as the one available in the online version (as represented in Figure 1 above). 36 Another difference between the online and offline data was the data restriction imposed on the downloaded corpus. In short, this meant that 5% of the corpus data were randomly selected and destroyed before download in order to ensure that the material would not infringe on the copyright of original news media publishers. The data destruction is described on the corpus website as a procedure in which “[e]very 200 words, ten words are removed and are replaced with “@””6. The corrupt data had therefore to be tracked and removed from the SQL-tables that would contain the corpus. Given the size of the corpus, this was, however, estimated to have only a marginal effect on the query results. 37 After the necessary arrangements were made in the SQL management environment, a random selection among the 2 000 most frequent common nouns in this part of NOW was performed which resulted in 100 lexical items constituting the base for truncated queries that were intended to be performed at a later stage. The limitation to the 2 000 most frequent nouns was motivated by the great number of items with little or no significance (for the present study) in the so-called long tail of infrequent forms. Examples of such items are presented in Table 1. The column PoS shows the grammatical function (i.e. Part-of-Speech) assigned by the automatic speech tagger CLAWS7 used by the NOW corpus. Lexis, 14 | 2019 213 Table 1: Long-tail forms excluded from the data. Wordform PoS Frequency Guild/AFTRA nn1 13 Liiiiittle nn1 2 Blues-- nn2 2 Dramatics nn1_nn2 2 WavSTAT nn1_vv0_np1 1 p-commerce nn1_jj 1 Chih-ping nn1 1 EYE/iStockPhoto nn1 1 $6,495 Nnu 1 LARIBA nn1_np1 1 38 The randomly selected 100 items were called candidate words (CaW) as they were intended to be the potential lexical sources in the retrieval of lexical blends. The first five CaWs in this list were testimony, track, prosecutor, routine, and citizen. The following step was to truncate the CaWs in order to prepare them for wildcard queries in the corpus. This truncation could not be done in an arbitrary fashion as the output from the CaWs would then depend on a variable beyond systematic control, i.e. comparisons would not be possible within the present dataset, nor with other datasets in potential future investigations. In other words, the truncation should be done according to a model employing the same rationale for all CaWs. The selected model for truncation needed therefore to be dependent on some identifiable and measurable variable in the corpus. To meet this end, the notion of selection point (SP) [Gries 2006] was chosen as it is a systematic identification of a breakpoint based on corpus frequency measures. 39 In short, the SP of a lexeme is located at the point at which a specific truncation generates a top frequency ranking of the CaW in question. Table 2 illustrates this with the CaW scope7. The point at which scope is truncated to scop* is, thus, where this truncated form renders scope as the most frequent item in the corpus query output (see fourth row in Table 2). The top frequency ranking of scope is, furthermore, taken to be a measure of its prominence in the data. To exemplify the actual output from the SP analysis, the ten truncated forms derived from the five first CaWs were testim*, *imony, trac*, *rack, pros*, *utor, routi*, *tine, citi*, and *zen. 40 The abbreviation wID in Table 2 is short for word ID, which is a categorization in the corpus roughly comparable to word type, although distinctions between wIDs include grammatical function (PoS), and is therefore more fine-grained than a traditional definition of type. Token frequency, however, is straightforwardly the number of items of Lexis, 14 | 2019 214 the wIDs in question found in the corpus. The column frequency rank shows the position of the CaW when the corresponding truncated form is employed in a wildcard query8, e.g. if the truncated form sco* is used in a query, the CaW scope is number four in the frequency ordered output list. In addition, it should be noted that even the entire CaW was subject to the truncation script query, which implies that no restriction was set on the possibility for an entire CaW to be the basis for further queries. Table 2: Calculation of selection point (cf. Gries [2006]) Truncation of wID Token Frequency Selected form for scope Frequency Frequency rank truncated queries s* 131 145 60 659 869 393 sc* 10 191 2 789 006 21 sco* 1 167 362 449 4 scop* 55 19 681 1 scop* scope* 28 19 445 1 41 Identifying the SP of the 100 CaWs (from both left and right) meant that the total number of truncated forms, henceforth referred to as truncforms, was 200. These were then used in corpus queries producing 200 lists of lexical items potentially containing some proportion of lexical blends. The entire data collection process is illustrated schematically in Figure 2. All steps from download of the database material to the queries based on the truncforms were automatic procedures following the predefined algorithm as described in this section. 42 Once the lists of lexical items generated from the last step (the truncform queries) were put together in word lists, the remaining analysis (see “Word lists resulting from corpus queries…” in Figure 2) was done manually. In the manual as well as the automated analyses, the specified limitations necessarily applied, which means that while the output was potentially productive, loss of data was inevitable. The strength of the procedure lies, however, in its transparency, which means that it can be replicated relatively easily, but also that contrasting datasets may provide further information in comparative studies. Lexis, 14 | 2019 215 Figure 2: The data collection process Image 1014BFF800005B60000036D0C9013DD13786D930.emf 3. Results 43 The output lists contained 33 256 lexical items (types) in total, which were analyzed manually. The wildcard query outputs based on all 200 truncforms ranged from 2 to 3 778 wordforms each, but the distribution of these was very uneven, which is illustrated in the mean output number of 166 wordforms, in comparison with the median number of 65. There were, in other words, a number of outliers that diverted significantly from the majority of the queries. For instance, the truncform testim* (from testimony) resulted in only 6 items in its respective wordlist, while dr* (from drive) generated 1 754 items. Notably, the number of blends in these lists were 1 and 6 respectively, which illustrates the unpredictability of the content in the lists. 44 Figure 3 displays a graph representing the total distribution of the 200 wildcard queries, and as a further illustration the 10 queries (out of 200, i.e. 5%) with the highest number of output constituted more than 44% of the total number of output word forms (14 720 of 33 256). In addition, there was a slight overweight (55%) of wordforms produced by right hand side truncforms, i.e. in which the final, or rightmost, segment of the CaW was removed (henceforth referred to as RT forms, e.g. scop*, as opposed to LT forms from which the initial, or leftmost, segment was removed, e.g. *cope). Lexis, 14 | 2019 216 Figure 3: Quantitative distribution of query outputs 45 The analysis of the output aiming at identifying lexical blends in the data employed the following rationale [Algeo 1977; Gries 2004a; Beliaeva 2014]. Query output items were considered lexical blends • if they included an initial truncated segment from SW1 and a final truncated segment from SW2, e.g. brunch (breakfast + lunch). • if only one of the SWs, either initial SW1 or final SW2, were truncated, e.g. webinar (web + seminar) or testimoney (testimony + money). • if none of the SWs were truncated but instead shared lexical material (overlap), e.g. stoption (stop + option). • if the blending of the SWs was non-linear but yet displayed truncation and/or overlap, e.g. Caleavefornia (California + leave). • if the orthography of at least one of the SWs was modified and one or all of criteria no. 1-4 still applied 46 The total number of lexical blends identified in the corpus output wordlists amounted to 78 items, or approximately 0.23% of the output. Importantly, this proportion should only be seen as a coefficient derived from the observed type frequency. Although its use is limited in this particular investigation, it may be a potentially powerful means to compare future datasets analyzed in a similar way. Furthermore, 46 of the 78 blends (59%) were found in lists generated from RT forms, which is partly explained as a result of the fact that there were already more of these wordforms than the ones generated by left hand side truncated CaWs (i.e. LT forms). 47 The above presented criteria for deciding what is a lexical blend were used consistently in the entire identification process, but as is shown in previous sections, drawing sharp categorical boundaries must necessarily involve decisions that include, as well as exclude, categorically unclear items [Brdar-Szabó & Brdar 2008]. 62 items in the data Lexis, 14 | 2019 217 were therefore put in a separate category for future reference, including forms such as petrodollar and dollywood. Although truncated according to a common blend pattern, petrodollar resembles rather a clipped modifier and a head, which seems more reasonable to interpret as an instance of compounding. Dollywood, on the other hand, may evoke the concept of Hollywood, but as the segment -wood is quite frequent in US place names (for instance Greenwood, Westwood, Englewood, Fleetwood, Glenwood, Wynwood, Stanwood) the overlap effect, or wordplay function [Renner 2015], was considered too marginal to include dollywood in the list of blends. Nonetheless, both these examples display the well-known difficulties to classify blends on the basis of their form [López Rúa 2004; Beliaeva 2014; Renner 2015]. 48 Employing such explicit criteria to classify blends may seem counterintuitive from a CL perspective. There are, however, two important points to be made in this respect. First, a distinction is made between membership and representativity in a category [Geeraerts 2006c]. In other words, while a decision needs to be made as to whether a lexical item is considered a lexical blend or not, actual instantiations display blend characteristics to a higher or lower degree. Second, the distribution of blend characteristics is not even. For instance, a semantic analysis of the blend relationshit (relationship + shit) has likely a humorous or satirical component, whereas its degree of structural overlap is moderate. In contrast, palimony (pal + alimony) is comparable in terms of degree of structural overlap, but its humorous potential is probably considered rather low. In Cognitive Linguistics, degree of representativity and uneven distribution are typically associated with the idea of family resemblance [Wittgenstein 1968; Rosch 1978; for a specific comment on blends see Brdar-Szabó & Brdar 2008]. Applied to the data of the present study, diverse items such as webinar (web + seminar), crooklyn (crook + Brooklyn), and herstorical (her + historical) are both considered lexical blends although they differ in terms of structure and meaning. For instance, while herstorical elegantly employs a non-morphematic breakpoint to highlight an unexpected aspect of the source word history (from a lexical point of view), webinar resembles a compound web seminar in analogy with blends such as motel without adding any significant level of wordplay and wittiness. 49 A notable observation in the context of the present study is that more than half of the identified lexical blends (42 out of 78) were not formed from the CaW that had generated them in the corpus queries. For instance, the RT query based on the CaW body (in which the segment bod- was applied) produced the lexical blend bodacious, commonly thought of as bold + audacious, thus following a well attested pattern of semantically related source words such as chill + relax in chillax and fantastic + fabulous in fantabulous [Algeo 1977; Kelly 1998; Gries 2012; Beliaeva 2014; Renner 2015]; i..e. there seemed to be no direct and necessary correspondence between the CaW body and the actual SW of the blend (i.e. bold). This may not strike as particularly surprising given the nature of the data collection procedure, but, at a closer look, these data are potentially informative. Notably, the segment used for each query was selected because it would primarily generate the CaW, at least on the basis of its frequency in the corpus. In the example of bodacious, there are attested sources of an ambivalence as to its first source word, i.e. bold vs. body [Prof. Maarten Lemmens, personal communication], but the frequency measures observed in the current study could explain why this is the case. Contextual analyses of bodacious in the data reveal that the otherwise rather semantically wide sources bold and audacious are narrowed metonymically in the blend so as to associate to aspects concerned with sexual categories. This is reflected in the Lexis, 14 | 2019 218 observation that the word body is quantitatively prominent, which is accounted for qualitatively as activation of a metonymically restricted conceptualization of body. In addition, the clipped version bod (from body) often used in colloquial American English further stresses the activation potential of the first syllable as body in bodacious [Prof. Raymond W. Gibbs, personal communication]. 50 Figure 4 illustrates the interplay between the lexical items body, bold, and audacious, which make up the conceptual basis for the blend bodacious. As the figure indicates, the conventional use of the blend is constrained by all three lexemes, despite the uncertain role of the lexeme body in the blend’s etymology. Figure 4: Metonymically derived associations in bodacious Image 101A6A5C0000661400003EB6D99C5D6831E46EAC.emf The pattern observed in bodacious is further exemplified in the blend compuware, which is a brand name of a large US based IT company9. Figure 5 is a graphical representation displaying the similarity in terms of semantic structure. The relation between the observed SW computer and the CaW company is, however, explicit in this case, which does not influence the structure per se, but instead adds to the information of the characteristics of the blend. Lexis, 14 | 2019 219 Figure 5: Metonymically derived associations in compuware Image 101A6A5C0000661400003EB691DD292C757B9355.emf 51 Among the blends with a discrepancy between CaW and SW were also forms such as crooklyn (crook + Brooklyn) RT derived from the CaW crowd. The data did not indicate any association between crook and crowd that would have allowed an interpretation in analogy with bodacious and compuware, which is not surprising given their different phonemic representations (/kɹaʊ-/ vs. /kɹʊ-/). Thus, the phonemic structure did not afford the specific semantic structure expressed in Figures 4 and 5. 52 An example of an intermediate case in this respect was eco-logical (ecological + logical) generated from the CaW local, which potentially involves a semantic link between environmental concerns and the benefits of the local community in terms of food production, sharing economy solutions, and travels10. Although the conceptual aspects of the blend were readily identifiable, the structural properties of eco-logical were considered too generic to support a straightforward semantic explanation of the bodacious and compuware type. The uneven distribution of characteristics noted concerning structural aspects is thus reflected in this property. Furthermore, while the conceptual ambiguity shares the uneven distribution of, for instance, structural overlap, there are no indications in the data of a correlation between these properties in terms of distribution. To illustrate with the examples above, bodacious displays a phonological overlap between the SWs (depending, however, on variation in pronunciation), whereas no overlap is identified in compuware. Eco-logical, on the other hand, which has some degree of conceptual ambiguity in terms of a relation between the CaW local and the blend, displays considerable SW overlap. Crooklyn is also clearly a blend relying on structural overlap, but its qualities as regards conceptual ambiguity between the blend and potentially contributing domains besides the SWs are indeed meagre. Indeed, the conventional understanding of family resemblance is clearly visible in these cases as the characteristics are varying in terms of degree at the same time as their distribution in the individual cases is uneven [Ungerer & Schmid 2006]. 53 Furthermore, approximately 9% of the blends (7/78) displayed a discrepancy between their phonemic and graphemic representation in the sense that the orthography was Lexis, 14 | 2019 220 necessary to distinguish the blend from one of its source words. Examples in the data were websight (website + sight) and robocup (robocop + cup). There were also examples of striking similarity between the blend and one of the SWs, such as in testimoney (testimony + money). Indeed, the similarity between the syllables /-moʊni/ and /- mʌni/ are so close that it cannot be precluded that the difference in rapid spoken conversation may potentially be confused. Therefore, although there were phonemic cues likely revealing the distinction between testimony and testimoney, the effect of similarity in the blend is hardly a random achievement (cf. Kelly [1998]; Gries [2004a]; Fábregas & Scalise [2012]). 54 The blend, and artist name [López Rúa 2010, 2012], Misstake (Miss + mistake) exemplified both patterns described above in that multiple possible meanings of miss are possible in the meaning construal of the blend, while at the same time the pronunciation alone does not reveal the blend as such. The overt multiple reference creates a richness of semantic potential that is not resolved in the blend, but instead retains ambiguity on several levels; source word origin, semantic potential, and tension between graphemic and phonemic representation. In addition, the powerful dynamics of the initial segment miss foregrounds the analyzability of the final segment -take, which is straightforwardly an English morpheme albeit not typically in the lexeme mistake. 55 As regards truncation patterns among the 78 blends, the quantitative prominence of RT forms in the output, which implies, in practice, that the first SW is truncated, was not reflected in the structure of the blends. Almost half of the blends (46%) were of the ABD type (e.g. webinar from web + seminar), i.e. forms in which SW1 was retained in full. The other half consisted of ACD and AD blends distributed relatively evenly, with only three examples of the A|BC|D type (palimony from pal + alimony, popera from pop + opera, and eco-logical from ecological and logical), in which truncation does not seem to follow the typical pattern. These proportions are represented in Figure 6. In other words, there was a marked tendency in the data towards preservation of the first source word, which is especially interesting since there was an output tendency going in the opposite direction. 56 It is also notable that the AD structure, which is traditionally seen as a prototypical truncation pattern in blends [Plag 2003; Gries 2004a; Brdar-Szabó & Brdar 2008] constituted only a fourth of the blends in the data. The centrality of an AD pattern in a prototype categorization account of blends need not necessarily be questioned for this reason, even if the findings of this investigation would prove robust in further studies. From a quantitative perspective, on the other hand, it seems problematic in relation to the observation in Beliaeva [2014: 33ff.] that “it is widely accepted […] that the majority of blends combine the initial part of the first source word (W1) with the final part of the second source word (W2); in terms of the formula […] AB + CD = AD”. Lexis, 14 | 2019 221 Figure 6: Distribution of truncation patterns 57 There were also examples of so-called intercalative blends in the data. These are blended forms “in which the two source words involved in the blend are so tightly integrated in the blended word that the sounds of one source lexeme are interspersed between the sounds of the other” [Kemmer 2003: 72]11. The ambiguity of such blends is recognized in Kemmer [2003], but in the context of the present study this characteristic is of particular interest. Examples from the data were dragula (drag + dracula), constraction (construction + action), testimoney (testimony + money), and egosystem (ego + ecosystem). The structure of the previously mentioned blends misstake, websight, and robocup was also of the intercalative type, although these items were more complex, and contained aspects that add to the intercalativity. 58 It was further noted that metonymy and metaphoricity were recurrent semantic characteristics in the formation of the blends in the data. In the conceptual domain web (in the metaphorically derived sense related to the Internet domain), the blend websight highlights seeing and being seen as an aspect of the role of a website, i.e. websites usually aim at attracting Internet users, which is typically profiled in visual categories. Importantly, this observation requires a familiarity with the Internet domain, especially since there is a literal sense of web that could, at least under certain circumstances, be a potential competitor SW. It follows from this that observing the figurative qualities depends (in this case perhaps apparently) to some extent on encyclopaedic knowledge in the construal of the blend’s semantic content. The figurativity of such blends signals thus the dependence on other types of knowledge than what has traditionally been thought of as linguistic knowledge. 59 Similarly, ethical considerations of the conceptual category testimony are selected as money is foregrounded in testimoney and gender inequalities in the designation of historical accounts are profiled as her is introduced in herstorical. The latter example Lexis, 14 | 2019 222 highlights, furthermore, the influence of creativity in relation to conventional morphological patterns. For reasons of space, the figurativity in lexical blending is not operationalized further in this study. Suffice it to say presently that figurativity is a neglected and potentially important semantic aspect in need of further investigation. 4. Discussion 60 The number of lexical blends that were identified in the 200 query output lists may appear rather low, especially given the numerous reports on the increasing popularity of blends [Bryant 1974; Kemmer 2003; Lehrer 2007; for a critical comment on this topic see also Lepic 2016]. Still, few studies have looked into this issue systematically enough to suggest clear definitions what is meant by increasing, let alone to present any figures behind such a presumably measurable change. Based on the findings of the present study only, it is therefore not possible, at this point, to say anything conclusive about the role of blends in natural language use, especially from a diachronic viewpoint. An investigation of a comparable dataset from the beginning of the 20th century (whether that would be obtainable at all) could possibly answer such questions, but as matters stand, this remains largely a field for speculation. Moreover, the limitations in terms of region and genre (especially the latter) have likely implications for the number of blends in the data of this investigation, which is in itself a call for further studies and replication of the current methodology on other datasets. 61 The problematic issue of defining what is a lexical blend (and not) further complicates the quantitative analyses [Bauer 2012]. One perspective that may contribute to developing the notion of category delineation is to take a prototype theory perspective [López Rúa 2004; Brdar-Szabó & Brdar 2008; Bauer 2012; Renner et al. 2012]. It follows from such a viewpoint that holistic analyses of the objects under investigation are performed, and that attributes, or characteristics, are studied in relation to their distribution among the data. The conventional way in Cognitive Linguistics is to interpret such relations in terms of family resemblance [Rosch 1978; Geeraerts 2006b; in the context of lexical blending see Brdar-Szabó & Brdar 2008], which enables comparisons with naturally occurring categories such as bird and dog. Methodological systematicity requires nonetheless that a strict distinction has to be made between degree of membership and degree of representativity. Membership in the category bird is discrete; something is or is not a bird. But some birds may be birdier than others: the swallow does remain a more typical bird than the ostrich. [Geeraerts 2006b: 150] 62 For the purposes of methodological design, definitions need therefore to be based on membership categorization, while analyses and discussions need to consider the degree of representativity as a central notion. In other words, no single study is likely to resolve the riddles of lexical blending. Instead, patterns emerging from a multitude of studies relying on systematic and replicable methodologies have better chances to increase the level of knowledge. 63 The analysis of the blend list revealed a few notable properties of the data as regards truncation. First, the majority of the blends (72% excluding A|BC|D blends) were formations in which only one of the source words was truncated. Second, although the initial queries resulted in significantly more word forms based on a right truncation (RT) of the CaWs, almost half of the blends were formed with the first source word Lexis, 14 | 2019 223 retained in full. Third, the widely spread idea that the AD structure should be the most prevalent pattern in lexical blending [Beliaeva 2014] was not supported by the data in this study, which is perhaps the most striking, and equally most problematic, observation of the truncation patterns. Furthermore, as news media (which is the origin of the data in the present study) have been claimed to be one of the central domains in the development of lexical blends [Bryant 1974; Lehrer 2007; Fandrych 2008] the expectation would have been that AD blends would have constituted a prominent part of the data. This was apparently not the case, and the observed discrepancy needs to be followed up by further investigations. Already at this point, however, it raises questions as to the nature of the data in previous studies. For instance, if no systematic rationale is employed in the data collection, it may be tempting to focus on powerful and witty blends, as these may seem to display a higher density of blend characteristics, and would therefore possibly be considered more representative. The blend misstake in the present study exemplifies such a formation employing several strategies, especially in terms of semantic content, and it may thus, on good grounds, be claimed to be highly significant as to qualitative description. This does not necessarily mean that it is representative of lexical blending as a process. Datasets that do not take this issue into account run therefore the risk to present lexical blending in a biased way. Systematic data collection seems thus ever so important in further accounts of this word formation process. The validity and the relevance of the findings are of course central notions in this respect, but perhaps equally important is the opportunity to ask questions about blends that have previously been difficult to address. 64 The observations of ambiguity in the blend data are in many senses parallel with the reasoning on wordplay in Renner [2015]; the emergence of various types of the phenomenon is more than haphazard consequences of certain formal structures. Although unevenly distributed in the data, both wordplay and ambiguity seem to be consistently, and functionally, employed as motivating attributes creating affordances for language users to form and use lexical blends. A preliminary division of the types of ambiguity found in the data in the present study can be described as i) source word ambiguity, ii) conceptual ambiguity, iii) mode ambiguity, and iv) truncation ambiguity. These attributes have the potential to co-occur freely in any instantiation, and the boundary between them is typically fuzzy. 65 First, source word ambiguity (e.g. dra- in dragula) have, for instance, been observed by Kemmer [2003] and Bauer [2012], but the functional properties of this characteristic have not yet been related to motivations for lexical blending. The ambiguous structure has rather been discussed as an analytical problem, which it is indeed from a structural perspective. The distinction between intercalative and non-intercalative structure is in fact questioned in Gries [2004b], where the apparent inconsistency is identified as the possibility for either analysis in Kemmer’s [2003] data. In contrast, it is argued in the present study that Kemmer [2003] has actually spotted an important motivation, or a cognitive constraint, for the formation and use of blends. Put differently, what has been perceived as a structural anomaly, or even an analytical flaw, is instead claimed here to be an important hint to the nature of lexical blending. 66 Second, the possibility for indirect sources to influence the meaning construal in blends such as bodacious and compuware constitutes another type of ambiguity, which profiles the activation of a third domain of meaning in blends that appear to have only Lexis, 14 | 2019 224 two sources. In the present study this characteristic is referred to as conceptual ambiguity. This type of ambiguity was relatively rare in the blend data, and further studies are needed to account for the distribution in detail. This strategy displays a striking similarity with figuratively motivated meaning construals as regards semantic structure; conceptual content is captured from an indirect domain, be it metonymically, metaphorically, polysemously, or even homonymously associated to the primary, or overt, source domains (SW1 and SW2). Given the common pattern of metonymically motivated combinations of concepts, the mechanism of conceptual ambiguity is not a surprising finding. 67 Third, the importance of phonology for lexical blending has sometimes been highlighted in previous research [Cannon 1986; Kubozono 1990; Lehrer 1996; Kemmer 2003]. Interestingly, the findings of the current investigation also underscore the orthography as a significant aspect of blend formation. While phonology is often related to structural concerns such as breakpoints, sound similarity, and prosody [Cannon 1986; Kelly 1998; Gries 2004a, 2006; Arndt-Lappe & Plag 2013], the functional use of spelling in the creation of blends seems to draw rather on semantic aspects, at least as regards the blends in the data of the present study. Moreover, the result of the discrepancy between orthography and phonology is understood as a dynamic interplay between these modes of language. The simultaneous activation of website and sight in the blend websight is thus the result of the synchronously produced ambiguity of the graphemic and phonemic representation, and the term chosen for this mechanism is subsequently mode ambiguity. 68 Finally, the blends palimony, eco-logical and popera illustrate a commonly observed pattern in lexical blending [Algeo 1977; Beliaeva 2014]. A conventional way to describe such items is to argue that the source words are retained in full in the blend (i.e. no truncation has occurred), and as this central characteristic is absent it may be tempting to exclude these items because they lack a central property (see for instance Ralli & Xydopoulos [2012]). This is questionable for at least two reasons. First, there is a growing understanding of lexical blends in terms of prototypically organized categories [López Rúa 2004; Brdar-Szabó & Brdar 2008; Bauer 2012; Renner et al. 2012; Renner 2015], which implies that identification and analyses relying on formal semantics listing of necessary and sufficient conditions are inherently misleading. It has even been argued that such approaches are (at least partly) responsible for the 20th century failures to account successfully for the complexity of lexical blending [Fandrych 2008]. Instead, holistic analyses of objects, including asymmetric attribute distribution, have proven much more apt to explain the complex nature of blends [Brdar-Szabó & Brdar 2008; Bauer 2012; Renner et al. 2012]. Thus, although truncation is seen as an important attribute of lexical blends, it does not have the status of a necessary condition, simply because blends cannot be categorized satisfactorily along such lines. 69 Second, the idea that both source words are retained in full in blends such as popera overlooks the structural and semantic realizations of ambiguity as a functional cognitive constraint. The analyses of the present study reveal several types of ambiguity as important motivations for lexical blending, and popera exemplifies one such ambiguity, albeit manifested in its structure rather than its semantics. It is therefore more reasonable to interpret its truncation as a structural ambiguity, or truncation ambiguity, produced as a result of the dynamicity between pop and opera as they are simultaneously represented in the blend. Put simply, while pop is profiled, Lexis, 14 | 2019 225 opera is backgrounded, and vice versa [Ungerer & Schmid 2006; Langacker 2008]. There is thus an apparent analogy with the well-established notion of binocular rivalry as a visual version of multistable perception [Levelt 1968; Rubin 2003] (Figure 7). Neither focus of attention allows a simultaneous attention to the contrasting object, which is realized in popera as a dynamically construed truncation, i.e. either of the SWs pop and opera are truncated depending on which of them is foregrounded. Figure 7: The face-vase illustration of binocular rivalry Image 1017EAD8000048E100004F3DA5D03CFABB24A5A4.emf 70 In all, one of the keys to separating lexical blending from other word formation processes may in fact be the functional use of ambiguity and uncertainty. From the perspective of similarity between SWs and between SWs and the blend, Gries [2012] indicates this by pointing out that if too much of the SWs are retained in blends, they are not likely to be successful simply because they are not fun anymore: while both source words are perfectly recognizable, the blend is not too similar to either source word anymore and the punning/playful character of such blends is largely lost. [Gries 2012: 159] 71 This reasoning on similarity is easily transferred to ambiguity as it is identified in the present study. As mentioned in the introduction, punning in this context is understood broadly as a concept drawing the attention to the idea of pleasure in the shape of intellectual challenges. The explanation for the impact of ambiguity, and indeed a motivation for the existence of lexical blends, is thus found in parallel with other intellectual challenges such as jigsaw puzzles, crosswords, and riddles. 72 The four types of ambiguity described in this section combine to a powerful cognitive constraint influencing, or even driving, the development and use of lexical blends; namely multistable meaning construal as an instantiation of the well attested cognitive Lexis, 14 | 2019 226 mechanism of multistable perception [Leopold & Logothetis 1999; Rubin 2003; Kornmeier et al. 2009]. Moreover, the underlying motivation for this constraint in blending is identified as a manifestation of a drive to seek intellectual challenge. 73 This explanatory frame encapsulates several findings of the blends in the data of the present study, but it is also in line with previous research, and moreover, it dovetails the Cognitive Linguistics claim that no sharp boundaries can be identified between language and other cognitive abilities and processes. Although lexical blends may be claimed to be marginal in comparison with word formation processes such as derivation and compounding, and more so yet in relation to lexical retrieval and syntactic processes in general, the findings in lexical blend research may have far- reaching implications for more general linguistic assumptions [Kelly 1998; Dressler 2000]. Also, outbursts of lexical creativity seen in controversial political circumstances [Lalić-Krstin & Silaški 2018] signal that the importance of lexical blends go far beyond the indications given by their low frequency numbers. Therefore, it seems even more pressing to take further steps towards a better understanding of these lexical items. 5. Conclusion 74 The two aims of the study – on the one hand an approach to data collection and methodology, and, on the other hand, an investigation of lexical ambiguity in the formation and use of blends – are in several ways indications for further research. The data collection methodology has provided important insights into the possibilities as well as difficulties in systematic collection of lexical blends. Some of the quantitatively oriented findings raise central questions about traditional conceptualizations of the nature of blends, while other observations need the context of further studies to be assessed as to their explanatory value. It seems clear, however, that Bauer [2012: 21] can be paraphrased successfully in the context of the present study by exchanging ill- defined with unsystematically collected in the phrase “[w]e need more than new experiments on an ill-defined set of words”. Bauer’s [2012] original intention in this quote contains still an unresolved issue, but systematicity is at the heart of further development not only as to general epistemological concerns, but also for the development of definitions and classificatory devices. As the present study shows, there is a potential in enriching the qualitative analysis with quantitative data in that previous conceptions of irregularity and complexity may be less dramatic in taxonomic matters than has been traditionally presumed. Also, it is demonstrated in this study how semantic explanations, such as the suggestion to understand ambiguity as an intentional process along the lines of multistable perception, have consequences for structural analyses as well. 75 The recurring claim that broad approaches are needed to explain lexical blending [Kelly 1998; Fandrych 2008; Renner et al. 2012] is further illustrated in this study. Categorization, classification and structural properties have been addressed in a multitude of accounts, but the semantic characteristics of blends constitute still a challenge to linguistic investigation. Methodological creativity is likely central in this respect. Kelly [1998] acknowledged this need two decades ago, and the current investigation hopefully adds another piece to the puzzle of lexical blending. Lexis, 14 | 2019 227 BIBLIOGRAPHY ALGEO John, 1977, “Blends, a Structural and Systemic View”, American Speech, 52, 47-64. ALM-ARVIUS Christina, 2012, Comprehensive Semantics. Lexicon, Grammar, Text Cognition & the World, Stockholm University: Unpublished manuscript. ARNDT-LAPPE Sabine & PLAG Ingo, 2013, “The Role of Prosodic Structure in the Formation of English Blends”, English Language and Linguistics, 17, 537-563. BAT-EL Outi, 2006, “Blend”, Encyclopedia of Language & Linguistics (Second Edition), 66-70, Oxford: Elsevier. BAUER Laurie, 1983, English word-formation Cambridge, New York: Cambridge University Press. BAUER Laurie, 2012, “Blends: Core and Periphery”, in RENNER V., ARNAUD P., GAST V. & MANIEZ F. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter: 11-22. BELIAEVA Natalia, 2014, “A Study of English Blends: From Structure to Meaning and Back Again”, Word Structure, 7, 29-54. BERGSTRÖM Gustaf Adolf, 1906, On Blendings of Synonymous or Cognate Expressions in English: A Contribution to the Study of Contamination, Lund University PhD Dissertation. BRDAR-SZABÓ Rita & BRDAR Mario, 2008, “On the marginality of lexical blending”, Jezikoslovlje, 9, 171-194. BRYAN Margaret M., 1974, “Blends are increasing”, American Speech, 49, 163. CANNON Garland, 1986, “Blends in English Word Formation”, Linguistics, 24, 723-753. COOK Paul & STEVENSON Suzanne, 2010, “Automatically identifying the source words of lexical blends in English”, Computational Linguistics, 36, 129-149. CRYSTAL David, 2001, Language and the Internet Cambridge, UK; New York: Cambridge University Press. CRYSTAL David, 2011, Internet Linguistics: a Student Guide New York: Routledge. DRESSLER Wolfgang U., 2000, “Extragrammatical vs. marginal morphology”, in DOLESCHAL U. & THORNTON A.M. (Eds.), Extragrammatical and marginal morphology, Munich: Lincom Europa: 1-10. FÁBREGAS A. & SCALISE S., 2012, Morphology: From data to theories, Oxford: Oxford University Press. FANDRYCH Ingrid, 2008, “Pagad, Chillax and Jozi: A Multi-Level Approach to Acronyms, Blends, and Clippings”, Nawa: Journal of Language & Communication, 2, 71-88. FRADIN Bernard, 2015, “Blending”, in MÜLLER P.O., OHNHEISER I., OLSEN S. & RAINER F. (Eds.), Word- Formation: an International Handbook of the Languages of Europe, Berlin/Boston: De Gruyter Mouton: 386-413. GEERAERTS Dirk (Ed.), 2006a, Cognitive Linguistics: Basic readings, New York: Mouton de Gruyter. GEERAERTS Dirk, 2006b, “Introduction: A rough guide to Cognitive Linguistics”, in GEERAERTS D. (Ed.), Cognitive Linguistics: Basic Readings, New York: Mouton de Gruyter: 1-28. GEERAERTS Dirk, 2006c, “Prototype Theory: Prospects and Problems of Prototype Theory” in D. GEERAERTS (ed.), Cognitive Linguistics: Basic Readings, New York: Mouton de Gruyter: 141-165. Lexis, 14 | 2019 228 GRIES Stefan Th., 2004a, “Isn’t that Fantabulous? How Similarity Motivates Intentional Morphological Blends in English”, in ACHARD M. & KEMMER S. (Eds.), Language, Culture, and Mind, Stanford, California: CSLI Publications: 415-428. GRIES Stefan Th., 2004b, “Shouldnt it be Breakfunch? A Quantitative Analysis of Blend Structure in English”, Linguistics, 42, 639-667. GRIES Stefan Th., 2006, “Cognitive Determinants of Subtractive Word Formation: A Corpus-based Perspective”, Cognitive Linguistics, 17, 535-558. GRIES Stefan Th., 2012, “Quantitative Corpus Data on Blend Formation: Psycho- and Cognitive- linguistic Perspectives”, in RENNER V., ARNAUD P., GAST V. & MANIEZ F. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter: 145-167. JUHASZ Barbara J., JOHNSON Rebecca L. & BREWER Jennifer, 2016, “An Investigation into the Processing of Lexicalized English Blend Words: Evidence from Lexical Decisions and Eye Movements During Reading”, Journal of Psycholinguistic Research, 1-14. KELLY Michael, 1998, “To ‘Brunch’ or to ‘Brench’: Some Aspects of Blend Structure”, Linguistics, 36, 579-590. KEMMER Suzanne, 2003, “Schemas and Lexical Blends”, in CUYCKENS H., BERG T. & DIRVEN R. (Eds.), Amsterdam Studies in the Theory and History of Linguistic Science. Motivation in Language: Studies in honor of Günter Radden, Amsterdam, NL: John Benjamins Publishing Company: 69-97. KJELLANDER Daniel, 2018, “Cognitive Constraints in English Lexical Blending: A Data Collection Methodology and an Explanatory Model”, Pragmatics & Cognition, 25, 142-173. KORNMEIER Jürgen, HEIN Christine Maira & BACH Michael, 2009, “Multistable Perception: When Bottom-Up and Top-Down Coincide”, Brain and Cognition, 69, 138-147. KUBOZONO Haruo, 1990, “Phonological constraints on blending in English as a case for phonology- morphology interface”, Yearbook of morphology, 3, 1-20. LALIĆ-KRSTIN Gordana & SILAŠKI Nadežda, 2018, “From brexit to bregret”, English Today, 34, 3-8. LANGACKER Ronald W., 2008, Cognitive Grammar: A Basic Introduction Oxford: Oxford University Press. LEHRER Adrienne, 1996, “Identifying and Interpreting Blends: An Experimental Approach”, Cognitive Linguistics, 7, 359-390. LEHRER Adrienne, 1998, “Scapes, holics, and thons: The semantics of English combining forms”, American Speech, 73, 3-28. LEHRER Adrienne, 2007, “Blendalicious” in MUNAT J. (Ed.), Lexical Creativity, Texts and Contexts, Amsterdam, NL: John Benjamins Publishing Company: 115-133. LEOPOLD David A. & LOGOTHETIS Nikos K., 1999, “Multistable phenomena: changing views in perception”, Trends in Cognitive Sciences, 3, 254-264. LEPIC Ryan, 2016, “Lexical blends and lexical patterns in English and in American Sign Language”, Paper presented to the Online Proceedings of the Tenth Mediterranean Morphology Meeting MMM10 7-10 September 2015, Haifa, 2016. LEVELT Willem J. M., 1968, On binocular rivalry, The Hague: Mouton. LÓPEZ RÚA Paula, 2004, “The Categorial Continuum of English Blends”, English Studies, 85, 63-76. Lexis, 14 | 2019 229 LÓPEZ RÚA Paula, 2010, “Making and breaking the rules: lexical creativity in the alternative music scene”, Language Awareness, 19, 51-67. LÓPEZ RÚA Paula, 2012, “Beyond all reasonable transgression: Lexical blending in alternative music”, in RENNER V., ARNAUD P., GAST V. & MANIEZ F. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter: 23-34. PLAG Ingo, 2003, Word-formation in English, Cambridge: Cambridge University Press. POUND Louise, 1914, “Blends: Their Relation to English Word Formation”, in HOOPS J. (Ed.), Anglistische Forschungen, Heidelberg: Carl Winter’s Universitätsbuchhandlung. RALLI Angela & XYDOPOULOS George J., 2012, “Blend formation in Modern Greek”, in RENNER V., ARNAUD P., GAST V. & MANIEZ F. (eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter: 35-50. RENNER Vincent, 2015, “Lexical blending as wordplay”, in ZIRKER A.& WINTER-FROEMEL E. (Eds.), Wordplay and Metalinguistic/Metadiscursive Reflection: Authors, Contexts, Techniques, and Meta- Reflection, Berlin, Boston: Walter de Gruyter GmbH: 119-133. RENNER Vincent, ARNAUD Pierre & MANIEZ François, 2012, “Introduction: A bird’s-eye view of Lexical Blending”, in RENNER V., ARNAUD P., GAST V. & MANIEZ F. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter: 1-9. RONNEBERGER-SIBOLD Elke, 2006, “Lexical blends: Functionally tuning the transparency of complex words”, Folia Linguistica, 40, 155-181. RONNEBERGER-SIBOLD Elke, 2012, “Blending between grammar and universal cognitive principles: Evidence from German, Farsi, and Chinese”, in RENNER V., ARNAUD P., GAST V. & MANIEZ F. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter: 115-143. ROSCH Eleanor, 1978, “Principles of Categorization”, in LLOYD B.B. & ROSCH E. (Eds.), Cognition and Categorization, Hillsdale, New Jersey: Lawrence Erlbaum Associates, Inc: 27-48. RUBIN Nava, 2003, “Binocular rivalry and perceptual multi-stability”, Trends in Neurosciences, 26, 289-291. SEBBA Mark, 2003, “Spelling rebellion”, in ANDROUTSOPOULOS J.K. & GEORGAKOPOULOU A. (Eds.), Discourse Constructions of Youth Identities, Amsterdam, Philadelphia: John Benjamins: 151-172. SHAW Philip, 2008, “Spelling, Accent and Identity in Computer-Mediated Communication”, English Today, 24, 42-49. SILAŠKI Nadežda & ĐUROVIĆ Tatjana, 2013, “Of ‘siliconaires’ and ‘millionerds’ - How ESP learners understand novel blends in English”, Iberica, 85-105. SOUDEK Lev, 1971, “The Development and Use of the Morpheme Burger in American English”, Linguistics, 9, 61-89. SPENCER A., 1996, Phonology, Oxford: Blackwell Publishing Ltd. TAYLOR John R., 2003, Linguistic Categorization, Oxford: Oxford University Press. UNGERER Friedrich & SCHMID Hans-Jörg, 2006, An Introduction to Cognitive Linguistics Harlow, UK: Longman, Pearson Education Ltd. WARREN Beatrice, 1990, “The Importance of Combining Forms”, in DRESSLER W., LUSCHUTZKY H., PFEIFFER O. & RENNISON J. (Eds.), Trends in Linguistics. Studies and Monographs [TiLSM]: Contemporary Morphology, Berlin/Boston: De Gruyter Mouton: 111-132. Lexis, 14 | 2019 230 WITTGENSTEIN Ludwig, 1968, Philosophical Investigations, Oxford: Blackwell. NOTES 1. This does not exclude the possibility for blends resulting from a speech error to be adopted in language use as instances of intentionally motivated lexical items. The term word formation blend merely indicates the use of an item regardless of its etymological origin. 2. https://www.english-corpora.org/now/ Retrieved 2019-05-14. 3. Structured Query Language (SQL) is a programming language designed to manage relational databases. 4. https://www.postgresql.org/ Retrieved 2019-05-14 5. https://www.pgadmin.org/ Retrieved 2019-05-14. 6. https://www.corpusdata.org/limitations.asp Retrieved 2019-10-15. 7. Table 2 in the present text is a variation on Gries’ [2006: 543] table 2, which uses the example word agitation. 8. The symbol ‘*’ is used here as it is the conventional way to signal wildcard queries. In SQL language, however, the wildcard query symbol is ‘%’. 9. https://www.compuware.com/ Retrieved 2019-08-19. 10. See for instance https://www.elmia.se/en/About-Elmia/Environment-and-safety/Eco- logical/ Retrieved 2019-08-20. 11. The alternative expression sandwich word to indicate this structure is mentioned in Algeo [1977]. ABSTRACTS The multifaceted and intermediary nature of lexical blending has been discussed from various theoretical perspectives in the last decades [Kubozono 1990; Kelly 1998; Dressler 2000; Kemmer 2003; Gries 2006; Fandrych 2008]. Corpus linguistic studies have contributed to a growing body of empirical data demonstrating significant patterns of blend formation [e.g. Gries 2004b, 2006; Beliaeva 2014]. These more recent findings are particularly important as they illustrate that lexical blending is not as irregular and unsystematic as it has often been assumed in the past [Lehrer 1996; Kelly 1998; Kemmer 2003; López Rúa 2004; Beliaeva 2014]. Because blends are often short-lived [Cannon 1986; Lehrer 1996] and informal [Bauer 1983; López Rúa 2010; Bauer 2012], making predictions about their nature is a complex task. Systematic analysis involves therefore a number of challenges. One such challenge is how to detect and collect these unspecified and un- tagged lexical items in a corpus. It is a well-known fact that conventional dictionaries are poor sources [Cannon 1986], and alternative methodologies are therefore required. Another challenge is how to ensure that the collected data is representative of all lexical blends within a selected set of limitations. Besides addressing these challenges, semantic aspects of ambiguity were investigated from a Cognitive Linguistics (CL) perspective [Geeraerts 2006a]. The patterns of ambiguity in the data are explained as instantiations of multistable perception, which is understood as a phenomenon in which “our perceptual system fails to produce a stable unambiguous percept” [Kornmeier et al. 2009: 138]. From a linguistic perspective, this means that Lexis, 14 | 2019 231 aspects such as phonology, orthography, and semantics constitute variables holding a potential for functionally employed lexical ambiguity [Renner 2015]. At various stages, the study employed both automated processes and manual analyses, which means that workload limitations applied to the scope of the data. A central aim of the study was, however, to develop systematic approaches to the collection of lexical blends, precisely because it is considered necessary in the development of blend research in general. La nature variée et le rôle intermédiaire de l’amalgame lexical ont été abordés selon diverses perspectives théoriques ces dernières décennies [Kubozono 1990 ; Kelly 1998 ; Dressler 2000; Kemmer 2003 ; Gries 2006 ; Fandrych 2008]. Les études en linguistique de corpus ont permis la constitution d’un ensemble grandissant de données empiriques qui mettent au jour des patrons productifs de formation des amalgames [Gries 2004b, 2006 ; Beliaeva 2014]. Ces découvertes récentes sont particulièrement intéressantes en ce qu’elles illustrent le fait que l’amalgame lexical n’est pas aussi irrégulier et asystématique qu’on l’a souvent présupposé par le passé [Lehrer 1996 ; Kelly 1998 ; Kemmer 2003 ; López Rúa 2004 ; Beliaeva 2014]. En raison du caractère généralement éphémère [Cannon 1986 ; Lehrer 1996] et informel [Bauer 1983 ; López Rúa 2010 ; Bauer 2012] des amalgames, il est compliqué de faire des prédictions quant à leur nature. Leur analyse systématique présente donc un certain nombre de défis. Un de ces défis est de savoir comment détecter et recueillir ces lexèmes non spécifiés et non étiquetés dans les corpus. Il est bien connu que les dictionnaires conventionnels sont une source insuffisante [Cannon 1986], et des méthodologies autres sont ainsi nécessaires. Un autre défi est de savoir comment s’assurer que des données recueillies sont représentatives de l’intégralité des amalgames lexicaux à l’intérieur d’un ensemble limité. La prise en considération de ces défis a été complétée par des études sur l’ambiguïté sémantique dans une perspective cognitive (Linguistique Cognitive) [Geeraerts 2006a]. Ces modèles d’ambiguité dans les données s’expliquent comme des instanciations d’une perception multistable, qui doit s’entendre comme un phénomène par lequel « our perceptual system fails to produce a stable unambiguous percept » [Kornmeier et al. 2009 : 138]. D’un point de vue linguistique, cela signifie que des aspects tels que la phonologie, l’orthographe et la sémantique constituent des variables à fort potentiel d’ambigüité lexicale utilisée de façon fonctionnelle [cf. Renner 2015]. À plusieurs reprises, cette étude a eu recours aussi bien à des procédés automatisés qu’à des analyses manuelles, ce qui signifie que des limitations quantitatives s’appliquent au champ des données recueillies. Cependant, un objectif principal de cette étude a été de développer une approche systématique en ce qui concerne la collecte des amalgames, précisément car cette question est considérée comme fondamentale dans le développement général des études sur les amalgames. INDEX Keywords: lexical blending, semantics, corpus linguistics, multistable perception, methodological development Mots-clés: amalgamation lexicale, sémantique, linguistique de corpus, perception multistable, développement méthodologique AUTHOR DANIEL KJELLANDER Umeå University, Sweden [email protected] Lexis, 14 | 2019 232 List of references Isabel Balteiro and Laurie Bauer 1 The following list of references includes works that are often cited in the studies dealing with blending in English. This bibliography cannot be exhaustive, as the subject is multifaceted and the studies then far too numerous to be listed. The following works or articles are thus suggested readings and other relevant references may of course be added. BIBLIOGRAPHY ALGEO John, 1977, “Blends, a Structural and Systemic View”, American Speech, 52, 47-64. ARNDT-LAPPE Sabine & PLAG Ingo, 2013, “The role of prosodic structure in the formation of English blends”, English Language and Linguistics 17,3, 537-563. BALTEIRO Isabel, 2013, “Blending in English Charactoons”, English Studies 94 (8), London and New York: Routledge, 883–907. BALTEIRO Isabel, 2018, “Emerging hybrid Spanish-English blend structures: Summergete con socketines”, Lingua 205, 1–14. BAT-EL Outi, 1996, “Selecting the best of the worst: the grammar of Hebrew blends”, Phonology 13, 283–328. BAT-EL Outi & COHEN Evan-Gary, 2012, “Stress in English blends: a constraint- based analysis”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), Cross-disciplinary perspectives on lexical blending, Berlin: De Gruyter Mouton, 193-212. BAT-EL Outi, 2006, “Blend”, in BROWN Keith (Ed.), Encyclopedia of Language & Linguistics, Second Edition, Vol. 2, Oxford: Elsevier, 66–70. Lexis, 14 | 2019 233 BAUER Laurie, 2012, “Blends: Core and Periphery”, in RENNER V., MANIEZ F. & ARNAUD P. (Eds.), Cross- Disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter: 11-22. BELIAEVA Natalia, 2014, “A study of English blends: From structure to meaning and back again”, Word Structure 7,1, 29–57. BELIAEVA Natalia, 2015, “Blends at the interface between compounding and clipping: Evidence from readers’ evaluations”, Neologica 9, 205–219. BELIAEVA Natalia, 2016, “Blends at the intersection of addition and subtraction: Evidence from processing”, SKASE Journal of Theoretical Linguistics 13,2, 23–45. BELIAEVA Natalia, 2019, “Blending in morphology”, in ARONOFF Mark (Ed.-in-chief), Oxford Research Encyclopedia of Linguistics, Oxford: Oxford University Press, available at https://oxfordre.com/ linguistics/abstract/10.1093/acrefore/9780199384655.001.0001/acrefore-9780199384655-e-511 BERGSTRÖM Gustaf Adolf, 1906, On Blendings of Synonymous or Cognate Expressions in English: A Contribution to the Study of Contamination, Lund University PhD Dissertation. BERMAN J.M., 1961, “Contribution on Blending”, Zeitschrift für Anglistik und Amerikanistik, 9, Germany: Walter De Gruyter, 278-281. BERREY Lester, 1939, “Newly-wedded words”, American Speech, 14, 1, 3-10. BRDAR-SZABÓ Rita & BRDAR Mario, 2008, “On the marginality of lexical blending”, Jezikoslovlje 9,1–2, 171–194. BRYANT, Margaret M., 1974, “Blends are increasing”, American Speech 49, 163-184. CANNON Garland, 1986, “Blends in English Word Formation”, Linguistics, 24, 725-753. CANNON Garland, 2000, “Blending”, in BOOIJ Geert, LEHMANN Christian & MUGDAN Joachim (Eds.), Morphology: An International Handbook on Inflection and Word-Formation, Berlin and New York: Walter de Gruyter, 952-956. CHOROLEEVA Kornelia, 2015, “Morpho-semantic groups of lexical blends”, Scientific Works Of University Of Food Technologies, 62, 902-904. CONNOLLY Patrick, 2013, “The innovation and adoption of English lexical blends”, JournaLIPP 2, 1-14. COOK Paul & STEVENSON Suzanne, 2010, “Automatically identifying the source words of lexical blends in English”, Computational Linguistics, 36, 129-149. FANDRYCH Ingrid, 2008, “Sub-morphemic elements in the formation of acronyms, blends and clippings”, Lexis, 2, 103-121. FAUCONNIER Gilles & TURNER Mark, 2002, The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities, New York: Basic Books. FRADIN Bernard, 2015, “Blending”, in MÜLLER P.O., OHNHEISER I., OLSEN S. & RAINER F. (Eds.), Word- Formation: an International Handbook of the Languages of Europe, Berlin/Boston: De Gruyter Mouton, 386-413. GRIES Stefan Th., 2004, “Isn’t that fantabulous? How similarity motivates intentional morphological blends in English”, in ACHARD Michel & KEMMER Suzanne (Eds.), Language, Culture, and Mind, Stanford, CA: CSLI, 415-428. GRIES Stefan Th., 2004, “Shouldn’t it be breakfunch? A quantitative analysis of the structure of blends”, Linguistics 42(3), 639-667. Lexis, 14 | 2019 234 GRIES, Stefan Th., 2004, “Some characteristics of English morphological blends”, in ANDRONIS Mary A., DEBENPORT Erin, PYCHA Anne, & YOSHIMURA Keiko (Eds.), Papers from the 38th Regional Meeting of the Chicago Linguistics Society: Vol. II. The Panels, Chicago, IL: Chicago Linguistics Society, 201-216. GRIES Stefan Th., 2006, “Cognitive determinants of subtractive word-formation processes: a corpus-based perspective”, Cognitive Linguistics 17, 4, 535-558. GRIES Stefan Th., 2012, “Quantitative corpus data on blend formation: psycho- and cognitive- linguistic perspectives”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), Cross- disciplinary perspectives on lexical blending, Berlin & New York: Mouton de Gruyter, 145-167. JUHASZ Barbara J., JOHNSON Rebecca L. & BREWER Jennifer, 2017, “An Investigation into the Processing of Lexicalized English Blend Words: Evidence from Lexical Decisions and Eye Movements During Reading”, Journal of Psycholinguistic Research, 281-294. KELLY Michael H., 1998, “To brunch or to brench: some aspects of blend structure”, Linguistics 36,3, 579-590. KEMMER Suzanne, 2003, “Schemas and Lexical Blends”, in CUYCKENS H., BERG T. & DIRVEN R. (Eds.), Amsterdam Studies in the Theory and History of Linguistic Science. Motivation in Language: Studies in honor of Günter Radden, Amsterdam, NL: John Benjamins Publishing Company, 69-97. KJELLANDER Daniel, 2018, “Cognitive Constraints in English Lexical Blending: A Data Collection Methodology and an Explanatory Model”, Pragmatics & Cognition, 25, 142-173. KUBOZONO Haruo, 1990, “Phonological constraints on blending in English as a case for phonology- morphology interface”, Yearbook of morphology, 3, 1-20. LEHRER Adrienne, 1996, “Identifying and Interpreting Blends: An Experimental Approach”, Cognitive Linguistics, 7, 359-390. LEHRER Adrienne, 1998, “Scapes, holics, and thons: The semantics of combining forms”, American Speech 73, 3–28. LEHRER Adrienne, 2007, “Blendalicious”, in MUNAT Judith (Ed.), Lexical Creativity, Texts and Contexts, Amsterdam/Philadelphia: Benjamins, 115-133. LÓPEZ RÚA Paula, 2004, “The Categorial Continuum of English Blends”, English Studies, 85, 63-76. LÓPEZ RÚA, Paula, 2012, “Beyond all reasonable transgression: Lexical blending in alternative music”, in RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), Cross-disciplinary perspectives on lexical blending, Berlin: De Gruyter Mouton, 23–34. MORETON Elliott, SMITH Jennifer L., PERTSOVA Katya, BROAD Rachel & PRICKETT Brandon, 2017, “Emergent positional privilege in novel English blends”, Language 93,2, 347‑380. NORDE Muriel & SIPPACH Sarah, 2019, “Nerdalicious scientainment: A network analysis of English libfixes”, Word Structure, Vol. 12 Issue 3, 353‑384. PIÑEROS Carlos-Eduardo, 2004, “The creation of portmanteaus in the extragrammatical morphology of Spanish”, Probus 16.2, 203-240. POUND Louise, 1914, Blends: Their Relation to English Word Formation, Heidelberg: Carl Winter’s Universitätsbuchhandlung. Lexis, 14 | 2019 235 RENNER Vincent, 2015, “Lexical blending as wordplay”, in ZIRKER A. & WINTER-FROEMEL E. (Eds.), Wordplay and Metalinguistic/Metadiscursive Reflection: Authors, Contexts, Techniques, and Meta- Reflection, Berlin, Boston: Walter de Gruyter GmbH: 119-133. RENNER Vincent, 2019, “French and English lexical blends in contrast”, Languages in Contrast, 19,1, 27-47. RENNER Vincent, ARNAUD Pierre & MANIEZ François, 2012, “Introduction: A bird’s-eye view of Lexical Blending”, in RENNER V., ARNAUD P., GAST V. & MANIEZ F. (Eds.), Cross-Disciplinary Perspectives on Lexical Blending, Berlin: De Gruyter: 1-9. RENNER Vincent, MANIEZ François & ARNAUD Pierre J.L. (Eds.), 2012, Cross-disciplinary perspectives on lexical blending, Berlin & New York: Mouton de Gruyter. RONNEBERGER-SIBOLD Elke, 2006, “Lexical blends: Functionally tuning the transparency of complex words”, Folia Linguistica 40, 1-2, 155-181. SILAŠKI Nadežda & ĐUROVIĆ Tatjana, 2013, “Of ‘siliconaires’ and ‘millionerds’ - How ESP learners understand novel blends in English”, Iberica, 25, 85-105. SOUDANEK L.J., 1998, “The Relation of Blending to English Word-Formation: Theory, Structure, and Typological Attempts”, in DRESSLER Wolfgang & MEID Wolfgang (Eds.), Proceedings of the Twelfth International Congress of Linguists: Vienna, August 28-September 2, 1977, Innsbruck: Institut für Sprachwissenschaft der Universität Innsbruck, 462-466. TOMASZEWICZ Ewa, 2008, “Novel words with final combining forms in English. A case for blends in word formation”, Poznań studies in contemporary linguistics 44.3, 363-378. AUTHORS ISABEL BALTEIRO Universidad de Alicante, Spain [email protected] LAURIE BAUER Victoria University of Wellington, New Zealand [email protected] Lexis, 14 | 2019 which is quite common in brand names; see Cook [2005]). Nor did we observe the use of foreign words that carry prestigious connotations, such as the French or Latin ones. As with any trend, it seems that non-standard spelling and foreign words have lost (some of) their allure in the meantime.