Stress and final /n/ deletion in Catalan: Combining Strict CV and OT Noam Faust, Francesc Torres-Tamarit

To cite this version:

Noam Faust, Francesc Torres-Tamarit. Stress and final /n/ deletion in Catalan: Combining Strict CV and OT. Glossa: a journal of general linguistics, Ubiquity Press, 2017, 2 (1), ￿10.5334/gjgl.64￿. ￿hal-01677686￿

HAL Id: hal-01677686 https://hal.archives-ouvertes.fr/hal-01677686 Submitted on 8 Jan 2018

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. a journal of Faust, Noam and Francesc Torres-Tamarit. 2017. Stress and final /n/ general linguistics Glossa deletion in Catalan: Combining Strict CV and OT. Glossa: a journal of general linguistics 2(1): 63. 1–23, DOI: https://doi.org/10.5334/gjgl.64

RESEARCH Stress and final /n/ deletion in Catalan: Combining Strict CV and OT Noam Faust and Francesc Torres-Tamarit CNRS/SFL (UMR 7023), Université Paris 8, 59/61, rue Pouchet, 75017 Paris, FR Corresponding author: Noam Faust ([email protected])

In Catalan, /n/ deletes in word-final position when it is preceded by a stressed vowel. In this paper, we present an analysis that combines the representations of Strict CV and the violable constraints of Optimality Theory, two rarely combined but perfectly compatible analytical tools. We propose that final /n/ floats unassociated below its C position if it cannot be licensed by a following segment. In such circumstances, an entire CV unit is left unidentified at the right edge, a situation which is penalized by the system. Still, when stress is final, this rightmost CV unit is identified by the weak branch of a trochaic foot. As a result, the /n/ may remain afloat. However, when stress is not final, the final CV remains otherwise unidentified, and so /n/ must associate to it, even though it is unlicensed.

Keywords: Catalan; Strict CV; Optimality Theory; stress; final /n/ deletion

1 Introduction All consonants in Catalan may appear in word-final position. However, two of them, /ɾ/ and /n/, can undergo a process of deletion in non-verbal forms.1 While the deletion of /ɾ/ is largely unpredictable, with a higher number of exceptions and items showing variation, that of /n/ is the general case in the native vocabulary of the language. A crucial aspect of the process is its conditioning factor: the application of final /n/ deletion requires the presence of stress on the vowel preceding the /n/ (1a). Indeed, /n/ does not delete if this condition is not met (2).2 In all cases, the presence of underlying /n/ can be retrieved from suffixed forms (1b, c) where, no longer word-final, the /n/ does surface.3 The process is productive, as evidenced by the words in (3a), which are all taken from the standard reference

1 Final /n/ deletion is sporadically found in verbal forms, as in [ˈbe]’ (s)he comes’ cf. [ˈbens] ‘you come’ or [̍ˈte] ‘(s)he has’ cf. [ˈtens] ‘you have’. However, most verbs show no final /n/ deletion, as in [ˈpɾεn] ‘(s)he takes’ or [ˈbεn] ‘(s)he sells’. We attribute final /n/ deletion in verbs to the same process found in nouns. The greater resistance of verbs to this process could be due to the higher degree of Paradigm Uniformity in verbal paradigms as compared to nominal ones. In this paper we concentrate on nominal elements, in which the process of final /n/ deletion is productive (with some lexical exceptions, which are dealt with in section 5). 2 There is a very small set of words (Mascaró 1976 cites eleven) in which final /n/ also underwent deletion after unstressed vowels, as in [ˈʒoβə] ‘young’ cf. [ʒuβəˈnisim] ‘very young’, [ˈɔmə] ‘man’ cf. [uməˈnεt] ‘small man’, and [ˈazə] ‘donkey’ cf. [əzəˈnεt] ‘small donkey’. In the plural, as opposed to the cases in (1b), /n/ also deleted: [ˈʒoβəs], [ˈɔməs], [ˈazəs]. We regard these cases as instances of diachronic deletion in which the scope of final /n/ deletion was not restricted to final /n/’s after stressed vowels. Evidence for the diachronic character of final /n/ deletion in these cases is the synchronic situation in other dialects of Catalan (Balearic and some Western dialects), in which /n/ does surface in the plural of such forms: [ˈʒoβəns], [ˈɔməns], [ˈazəns]). See Jiménez and Lloret (2011) for an accurate description of the dialectal variation found in Catalan with respect to the extent of final /n/ deletion. 3 For a diachronic approach to final /n/ deletion in Catalan, see Gulsoy (1997) and Sampson (1999). Art. 63, page 2 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

dictionary of Catalan (except for sincrotró ‘synchrotron’). We return to the issue of the productivity of the process in loan and learned words in section 3.3.6. Most of the data in this paper is taken from Mascaró (1976).

(1) Final /n/ deletion in Catalan a. Singular b. Plural c. Derivative [ˈpla] [ˈplans] [pləˈnisim] ‘even’ ‘very even’ [kuˈzi] [kuˈzins] [kuziˈnεt] ‘cousin’ ‘small cousin’ [∫iˈlɛ] [∫iˈlεns] [∫iləˈnizmə] ‘Chilean’ ‘Chileanism’ [upurˈtu] [upurˈtuns] [upurtuniˈtat] ‘opportune’ ‘opportunity’ [distɾiβuˈsjo] [distɾiβuˈsjons] [distɾiβusjuˈnal] ‘distribution’ ‘distributional’

(2) Blocking of final /n/ deletion in words with non-final stress a. Antepenultimate b. Penultimate [əsˈpεsimən] [ˈatun] [əɡˈzamən] [ˈkɾεmlin] ‘specimen’ ‘atonic’ ‘exam’ ‘Kremlin’ [iˈpεrβətun] [‘platən] [uˈɾiƷən] [ˈramən] or [ˈramen] ‘hyperbaton’ ‘banana’ ‘origin’ ‘ramen noodles’

(3) Final /n/ deletion in loanwords a. Singular b. Plural c. Derivative [uˈzo] [uˈzons] [uˈzɔnik] ‘ozone’ ‘ozonic’ [təliˈβa] [təliˈβans] [təliβəˈnizmə] ‘Taliban’ ‘Talibanism’ [rəməˈða] [rəməˈðans] [rəməðəˈnεt] ‘Ramadan’ ‘small Ramadan’ (Twitter) [siŋkɾuˈtɾo] [siŋkɾuˈtɾons] ‘synchrotron’

Nevertheless, one does find word-final nasals preceded by a stressed vowel. These are found under one of three conditions: (i) if [n] is the second consonant of a final /rn/ cluster (4a); (ii) if [n] is the first consonant of a nasal-stop cluster in word-final position that has been simplified by deleting the last consonant (4b); (iii) in a list of lexical exceptions within the native vocabulary (4c); and in many loan and learned words (4d) (cf. the loanwords in (3)), although substandard forms in which the /n/ deletes are found, as will be shown later. Note that condition (ii) can result in the appearance of a final velar nasal. Labial and palatal nasals in coda position never delete (4e). Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 3 of 23

(4) No final /n/ deletion a. /rn#/ b. /nC#/ [ˈkarn] [kəˈlen] cf. [kəˈlentə] ‘meat’ ‘hot.m’ ‘hot.f’ [kunˈtorn] [ˈblaŋ] cf. [ˈblaŋkə] ‘contour’ ‘white.m’ ‘white.f’ c. Lexical exceptions d. Loanwords e. Final [m, ɲ] [ˈsɔn] [məɣəˈzin] [ˈam] ‘sleep, ‘magazine’ ‘fish hook’ sleepiness’ [ʒuˈan] [uɾəŋɡuˈtan] [əksˈtɾem] ‘John’ ‘orangutan’ ‘extreme’ [ˈnεn] [∫ərləsˈton] [∫əmˈpaɲ] ‘boy’ ‘Charleston’ ‘champagne’ [əˈran] [∫əˈman] [ˈboɲ] ‘close’ ‘shaman’ ‘bump’

If so, it seems that Catalan usually avoids [n] in word-final coda position. This is curious because cross-linguistically nasals are ideal coda consonants; [n] is a sonorant and has default place of articulation (a claim we motivate below for Catalan). If a language allows coda consonants, it is very likely to have nasal codas.4 In light of this, one may ask why in Catalan, of all the consonants that may appear in word-final position, /n/ undergoes synchronic deletion. Indeed, this process is the opposite of what one would expect from the point of view of markedness: labial and palatal nasals are allowed in coda position, as are all non-coronals, but coronal nasals are banned from this position. In addition, final stress in Catalan is most commonly found in consonant-final words (Serra 1996; Vallverdú 1997; Bonet & Lloret 1998). The deletion of final /n/ renders this metrical generalization non-surface true, that is, opaque. For all these reaons, Bonet et al. (2004) call /n/ deletion a “crazy” rule. In this paper, we show that there is nothing “crazy” about final /n/ deletion. In fact, final /n/ deletion is only expected once we formalize the reason why /n/ is such a good coda in Catalan. In addition to the motivation for final /n/ deletion (Q1 below), the analysis also provides answers to questions 2–5 in (5).

4 See VanDam (2004) and Blevins (2004, 2006) on the typology of word-final coda consonants in the world’s languages. Both authors coincide in showing that nasals are among the most common sounds in word-final coda position. The next quote is from Blevins (2006): “Sonority has also been taken to play a role in limiting single member onsets to low sonority segments, and single member codas to high sonority segments. However, there is ultimately no secure empirical basis for such generalizations (Blevins 2004). No language limits onsets to obstruents, nor does any known language ban all sonorants from onset position. Although there are some languages in which single codas are limited to particular sonorants, these patterns do not cover contiguous ranges on the sonority scale. For example, in Manam, and Oceanic language, where only nasals are possible codas, liquids l and r must be prohibited from coda position, which raises problems for a sonority-based account. See Blevins (2004), where convergent evolution is shown to account for the preponderance of nasal codas in the world’s languages”. Art. 63, page 4 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

(5) Research questions Q1. Why is /n/ the only nasal coda to be systematically deleted? And why only word-finally? Q2. Why must stress immediately precede the deleted /n/? Q3. Why is final /n/ deletion blocked when /n/ is the second element in a consonant cluster? Q4. Why does final cluster simplification counterfeed final /n/ deletion? Q5. What is the status of lexical exceptions?

The analysis makes use of two theories that have rarely been combined, although they are completely compatible: the representations employed are those of Strict CV (Lowenstamm 1996; Scheer 2004), whereas the computation is formalized within Optimality Theory (Prince & Smolensky 1993/2004). (For previous attempts at combining these two theories see, for instance, Polgárdi 1998 and Rowicka 1999). The outline of the paper is as follows. In section 2, we briefly present the relevant aspects of Government Phonology and Strict CV. We also provide the basic facts and analysis of stress assignment in Catalan nominal elements. In the analysis in section 3, it is proposed that /n/ needs to be licensed by a following segment in order to be associated. Word-finally, this is not possible because there is no licensor to its right. This non-association creates the problem of an unidentified CV unit. However, when stress is final, the segmentally unidentified CV structure is nevertheless engaged by the metrical structure. If so, only when stress is not final does the final CV remain unidentified at all levels; and only in those cases must the /n/ be associated to its C position even though the nasal is unlicensed. The analysis is also extended to the rest of the data, that is, to those cases in which final /n/ does not delete because it is the second element of a final /rn/ cluster or because it is the first element of an /nC/ cluster, and to lexical exceptions. Section 4 includes a brief discussion of previous approaches, and the paper concludes with section 5. 2 Background 2.1 Government Phonology and its Strict CV version In this subsection, we introduce Government Phonology (henceforth GP, Kaye et al. 1990) and its offspring Strict CV (Lowenstamm 1996; Scheer 2004). The introduction is very brief, focusing on the aspects of GP and Strict CV that are relevant to our purposes.

2.1.1 Government Phonology GP is principally a theory of autosegmental representations. Its main tenet is the existence of empty positions in the syllabic structure. The theory’s starting point is that the phonetic adjacency of two sounds may realize more than one phonological structure. Three possible structures are proposed for two adjacent consonants: (i) branching onset; (ii) coda-onset; and (iii) onset-onset. To illustrate, consider the case of French, where the sequence /tʁ/ has a similar distribution to that of a simple onset: it can appear word-initially [tʁuv] ‘find’ or word- medially after a consonant [mistʁal] ‘Mistral wind’. Such clusters are thus represented as branching onsets (6a). The inverse cluster /ʁt/, on the other hand, is never found word- initially or word-medially after a consonant. In that case, /ʁ/ is therefore syllabified as a coda, [paʁti] ‘party’ (6b). Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 5 of 23

(6)(6) BranchingBranching onset onsetvs. vs.coda-onset coda-onset cluster cluster in GP in GP a. O R O R b. O R O R N N N N (6) Branchingx x onsetx x vs.x coda-onsetx clusterx x inx GPx a. O R O R b. O R O R p a t K i p a K t i N N N N Codas are managed by the principle of coda licensing (Kaye 1990), which states that codas must be licensedx byx ax followingx x onset. Thisx x relationx x ofx licensing is represented in (6b) by an Codas are managed by the principle of coda licensing (Kaye 1990), which states that arrow. The principlep a explainst K i the impossibilityp a ofK *VCt KitV: an onset can license only one pre- cedingcodas consonant,must be licensed and so no by other a following consonant onset. can precede This therelation coda. of The licensing same principle is represented has anotherinCodas (6b) effect, areby managedan namely arrow. by that The the final principleprinciple consonants of explainscoda cannot licensing bethe codas, impossibility(Kaye because 1990), they whichof *VCʁtV: are states not licensed thatan onset codas in can thislicensemust position. be only licensed one This bypreceding is a followingwelcome consonant, result, onset. since This and relation there so no is of muchother licensing empirical consonant is represented support can precede for in final (6b) con- bythe an coda. sonantsThearrow. same behaving The principle principle as onsets explains has (Harris another the impossibility & Gussmannimplication, of 1998, *VC namely 2002).KtV: an that Lastly, onset final can if final license consonants consonants only one are pre-cannot be incodas,ceding fact onsets, because consonant, then they the and are final so not no constituent otherlicensed consonant of in the this word can position. is precede an empty This the nucleus. coda.is a welcome The The same consonant-final result, principle since has there Catalanisanother much word effect,empirical[Sop] namely‘drenched’ support that final isfor thus consonantsfinal represented consonants cannot as in be (7).behaving codas, because as onsets they are(Harris not licensed & Gussmann in 1998,this position. 2002). Lastly, This is aif welcomefinal consonants result, since are there in isfact much onsets, empirical then support the final for finalconstituent con- of sonants(7) Final behaving consonant as onsets as onset (Harris of an & empty Gussmann nucleus 1998, in GP 2002). Lastly, if final consonants are the word is an empty nucleus. The consonant-final Catalan word [∫op] ‘drenched’ is thus in factO onsets,R O thenR the final constituent of the word is an empty nucleus. The consonant-final represented as in (7). Catalan word [Sop] ‘drenched’ is thus represented as in (7). N N (7) (7) FinalFinal consonantconsonant as as onset onset of anof emptyan empty nucleus nucleus in GP in GP xO xR xO xR

S oN p N

x x x x

S o p The concept of licensing, it must be added here, is not limited to coda-onset clusters. Rather, it has been proposed that an analogous relation exists between a realized nucleus and its onset, theThe nucleus concept licensing of licensing, the preceding it must onset. be5 added here, is not limited to coda-onset clusters. Rather, it has been proposed that an analogous relation exists between a realized nucleus 5 2.1.2andThe its Strict concept onset, CV of the and licensing, nucleus the licensing it licensing must conditions be added the preceding here, for nasal is not ‘codas’onset. limited to coda-onset clusters. Rather, it has been proposed that an analogous relation exists between a realized nucleus and its onset, In2.1.2the this nucleusStrict paper, CV licensing we and adopt the thelicensing a more preceding radical conditions onset. view5 offor GP, nasal namely ‘codas’ the Strict CV version proposed in LowenstammIn this paper, (1996) we adopt and developeda more radical in Scheer view (2004). of GP, Innamely this version, the Strict the CV skeleton version and proposed the syllabicin2.1.2 Lowenstamm Strict constituents CV and (1996) of the GP licensing areand collapsed developed conditions into onein for Scheer flat nasal sequence ‘codas’(2004). of CVIn units;this version, CV is in the fact theskeleton onlyand the skeletal syllabic unit recognizedconstituents by theof GP theory. are collapsed In order to into integrate one branchingflat sequence onsets of and CV coda units; CV In this paper, we adopt a more radical view of GP, namely the Strict CV version proposed in licensingis in fact inthe this only version skeletal of GP, unit we recognized assume, together by the with theory. Scheer In (2004), order to that integrate two segments branching mayLowenstamm entertain relations (1996) and that developed rid the grammar in Scheer of (2004).the need In to this take version, care of the intervening skeleton empty and the onsets and coda licensing in this version of GP, we assume, together with Scheer (2004), nuclei.syllabic Thus, constituents the segments of GP of are the collapsed branching into onset one flat in Spanish sequence["blaNko] of CV units;‘white. CVMASC is in’fact in (8) the that two segments may entertain relations that rid the grammar of the need to take haveonly an skeletal empty V-slotunit recognized between them, by the but theory. because In of order the infrasegmental to integrate branching relation onsetssymbolized and coda by thecarelicensing arrow of intervening inthis this empty version empty slot of does GP,nuclei. not we pose assume, Thus, a problem. the together segments with Scheerof the (2004),branching that twoonset segments in Spanish [‘blaŋko]may entertain← ‘white. relationsmasc that’ in rid(8) thehave grammar an empty of the V-slot need between to take care them, of intervening but because empty of the 5This of course raises the issue of how the final onset is licensed. Standard GP assumes a parametric approach infrasegmentalnuclei. Thus, the relation segments symbolized of the branching by the onset arrow in Spanish ← this["blaNko] empty‘white. slot doesMASC not’ in (8) pose a to this issue: final empty nuclei either license the preceding onset or not. A parameter like this can be easily formalizedproblem.have an as empty the result V-slot of constraint between interaction them, but in because OT. With of respect the infrasegmental to coronal nasals, relation however, symbolized the final empty by nucleusthe arrow does not providethis empty place, slot and doestherefore not the pose nasal a problem.is still unlicensed in this position. ← 5This of course raises the issue of how the final onset is licensed. Standard GP assumes a parametric approach to this issue: final empty nuclei either license the preceding onset or not. A parameter like this can be easily 5 This of course raises the issue of how the final onset is licensed. Standard GP assumes a parametric approach formalized as the result of constraint interaction in OT. With respect to coronal nasals, however, the final empty to this issue: final empty nuclei either license the5 preceding onset or not. A parameter like this can be easily nucleusformalized does as not the provide result place, of constraint and therefore interaction the nasal isinstill OT. unlicensed With respect in this to position. coronal nasals, however, the final empty nucleus does not provide place, and therefore the nasal is still unlicensed in this position.

5 Art. 63, page 6 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

(8) Branching onset in Strict CV (8) (8) BranchingC V C onset onsetV C in in StrictV StrictC CVV CV C V C V C V C V b l a N k o b ← l a N k o ←

The word in (8) illustrates a well-known cross-linguistic empirical fact, namely that nasals assimilate in place of articulation to a following obstruent. This fact was expressed in The word in (8) illustrates a well-known cross-linguistic empirical fact, namely that nasals standardTheassimilate word GP in through place (8) illustrates of articulationcoda alicensing well-known to a following and cross-linguisticthe adjacency obstruent. This empiricalof the fact coda was fact, and expressed namely onset inthatpositions. standard nasals But inassimilateGP Strict through CV, in all coda place sonorant-obstruent licensing of articulation and the to a adjacency clusters following are of obstruent. thein fact coda separated This and fact onset was by positions. expressedan empty But in nucleus. standard in Strict How canGPCV, the through all homorganicity sonorant-obstruent coda licensing of clustersnasal-obstruent and the are adjacency in fact clusters of separated the coda be by accounted and an onsetempty positions.for? nucleus. Strict HowBut CV in canrecognizes Strict the twoCV,homorganicity lateral all sonorant-obstruent forces: of nasal-obstruent government clusters and clusters are licensing. in befact accounted separated Both operate for? by an Strictempty from CV right nucleus. recognizes to left. How two Government can lateral the inhibitshomorganicityforces: government the pronunciation of nasal-obstruent and licensing. of segments clusters Both operate be while accounted from licensing right for? toStrict allows left. CV Government forrecognizes their pronunciation. inhibits two lateral the forces: government and licensing. Both operate from right to left. Government inhibits the Nowpronunciation consider ofthe segments nasal in while (9). licensingV3 governs allows the for preceding their pronunciation. empty nucleus Now considerV2 (the thedashed arrow)pronunciationnasal in and (9). licenses Vof3 segmentsgoverns its immediately the while preceding licensing precedingempty allows nucleus for onset. their V2 pronunciation. (the The dashed empty arrow) nucleus, Now considerand in licenses contrast, the doesnasalits immediately not in have (9). Vany3 precedinggoverns lateral the onset.force, preceding and The thus empty empty cannot nucleus, nucleus license in V contrast,2 (thethe preceding dashed does notarrow) nasal. have and any The licenses lateral position thatitsforce, immediatelythe and nasal thus occupies cannotpreceding license is onset. therefore the The preceding emptyweak. nucleus,nasal.Due to The inthis contrast, position weakness,that does the notthe nasal havenasal occupies any “pirates”, lateral is in force, and thus cannot license the preceding nasal. The position that the nasal occupies is Scheer’stherefore (2004) weak. Dueterms, to thisthe weakness,feature of the the nasal following “pirates”, obstruent. in Scheer’s The (2004) result terms, is a thekind feature of “nasal therefore weak. Due to this weakness, the nasal “pirates”, in Scheer’s (2004) terms, the feature geminate”,of the following in which obstruent. the primes The result or features is a kind of ofthe “nasal obstruent geminate”, are shared in which by the primesposition or that offeatures the following of the obstruent obstruent. are The shared result by is the a positionkind of “nasal that the geminate”, nasal occupies. in which Under theprimes this view, or the nasal occupies. Under this view, nasal homorganicity in Strict CV is a consequence not featuresnasal homorganicity of the obstruent in Strict are shared CV is a by consequence the position not that of the adjacency, nasal occupies. as in Standard Under GP, this but view, of of adjacency, as in Standard GP, but of positional weakness. nasalpositional homorganicity weakness. in Strict CV is a consequence not of adjacency, as in Standard GP, but of positional weakness. (9) (9) StructureStructure resulting resulting from from homorganization homorganization (Scheer (Scheer 2004: 2004: 709) 709) (9) Structure resulting from homorganization (Scheer 2004: 709)

V1 C V2 C V3 V1 C V2 C V3 V α V V α V N β N β γ γ In the final position, however, a weak nasal cannot be saved through place sharing, because InInthere the the is final final no consonant position, position, however, after thehowever, nasal. a weak As nasal aa consequence, cannotweak nasal be saved another cannot through process placebe applies saved sharing, in nasal through because neu- place sharing, becausetheretralizing is no there languages, consonant is no namely after consonant the lenition, nasal. after whichAs a it. consequence, consists As a consequence, of setting another the process nasal another place applies process specification in nasal applies neu- to in nasaltralizingtheir unmarkedneutralizing languages, value languages, namely (coronal lenition, or namely velar which depending lenition, consists on which of the setting language). consists the nasal6 of placesetting specification the nasal toplace 6 specificationtheirIn unmarked Catalan, to valuemedial their (coronal pre-consonantalunmarked or velar value depending/n/ (coronalassimilates on theor invelar language). place depending of articulation on the to language). the follow-6 In Catalan, medial pre-consonantal assimilates in place of articulation to the follow- ingIn consonant, Catalan, medial and thus pre-consonantal a place-sharing/n/ configuration /n/ assimilates like the in one place presented of articulation above suffices. to the ing consonant, and thus a place-sharing configuration like the one presented above suffices. followingWord-final consonant,/n/, however, and takes thus a a radical place-sharing turn. As the configuration weakness of the like nasal the in one word-final presented posi- above Word-finaltion cannot/n/ be, rescued however, by takes a sharing a radical configuration, turn. As the which weakness would of license the nasal the in nasal, word-final it is simply posi- suffices. Word-final /n/, however, takes a radical turn. As the weakness of the nasal in tiondeleted. cannot In the be rescued sections by devoted a sharing to the configuration, formal analysis, which we would are more license explicit the nasal, about it the is process simply word-final position cannot be rescued by a sharing configuration, which would license deleted.of /n/ deletion, In the sections and why devoted word-final to the labial formal nasals analysis, do not delete, we are just more like explicit coronal about and velarthe process nasals theofthat /n/nasal, comedeletion, it to is be simply and word-final why deleted. word-final due to In final labialthe consonantsections nasals do devoted cluster not delete, simplification. to justthe likeformal coronal For analysis, now and it velar suffices we nasals are to more explicitthatsay that come about in to Catalan, be the word-final process only the dueof presence /n/ to final deletion, of consonant a consonant and cluster why can word-final simplification. license a nasal labial For if it nownasals occupies it sufficesdo anot weak todelete, justsayposition, like that coronal in i.e. Catalan, when and not only velar followed the nasals presence by athat full of come a V consonant that to licenses be word-final can it. license a nasaldue to if final it occupies consonant a weak cluster simplification.position, i.e. when For not now followed it suffices by a full to V say that that licenses in Catalan, it. only the presence of a consonant 6 can licenseScheer (2004) a nasal proposes if it that occupies the syllabic a statusweak of position, final nasals ini.e. Germanic when isnot driven followed by the same by force. a full V that 6Scheer (2004) proposes that the syllabic status of final nasals in Germanic is driven by the same force. licenses it. A final issue concerns final empty6 nuclei (FEN). If empty nuclei are inhibited by 6 government, what inhibits the FEN in Catalan [∫op] above? In GP, FEN are parametrically

6 Scheer (2004) proposes that the syllabic status of final nasals in Germanic is driven by the same force. Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 7 of 23

licensed. For our purposes, we may simply say that FEN’s are not governed, but remain unrealized for a different reason. We have introduced all the concepts of Strict CV that are necessary for the ensuing analysis. Let us now summarize them in (10).

(10) Summary of Strict CV concepts i. The skeleton and syllabic constituents are conflated into a sequence of CV units. ii. Final consonants occupy an onset position followed by an empty nucleus. iii. Assimilated preconsonantal nasals are licensed by place sharing above an intervening empty nucleus. iv. Word-final nasals cannot be licensed because they are not followed by any consonant; therefore, they undergo different processes of lenition, or even deletion as in Catalan (which can be understood as an extreme case of lenition).

With these tools in hand we may turn to a description of the Catalan stress system in the next subsection. 2.2 Stress in Catalan In Catalan, nominal forms have either regular or lexical stress. Regular stress falls on the last vowel if the word is consonant-final (11a) and on the penultimate vowel if the word is vowel-final (11b).7 Words with final /n/ deletion behave like consonant-final words in this respect (11c). Most words ending in a non-alternating [n] (11d) behave like irregularly stressed words, that is, consonant-final words with penultimate stress. (Here we must exclude the exceptional non-alternating cases in (4c, d) with final /n/ and immediately preceding stress). Lexical stress can appear on any vowel within a right-oriented three- syllable window. Penultimate stress is exceptional in consonant-final words (12a), final stress is exceptional in vowel-final words (12b), and there are also words ending in a vowel or a consonant that show antepenultimate stress (12c). The examples in (11) and (12) contain both monomorphemic and derived words. Morphological structure is in most cases irrelevant for stress assignment.8

(11) Stress in Catalan a. [-C#] b. [-V#] c. /-n#/ d. [-n#] [əlməˈnak] [nəˈɣɔsi] [kəˈmi] [miˈzɔʒin] ‘almanac’ ‘business’ ‘path, way’ ‘misogynist’ [pəˈtit] [kəˈnaɾi] [sulˈta] [ˈkɾεmlin] ‘small’ ‘canary’ ‘sultan’ ‘Kremlin’ [nəsjuˈnal] [munˈtaɲə] [urɣənidzəˈsjo] [ˈkɔlun] ‘national’ ‘mountain’ ‘organization’ ‘colon’ [bunˈdat] [biˈɣɔti] [əməɾiˈka] [əβuˈɾiʒən] ‘goodness’ ‘mustache’ ‘American’ ‘aboriginal’

7 The plural morph [-s] does not trigger final stress. 8 There is in Catalan a set of derivational suffixes that are stress-attracting but that must remain unstressed, as in [ənˈʒεlik] ‘angelical’ cf. [ˈanʒəl] ‘angel’ (Mascaró 1976; Wheeler 2005) (see also the examples in (12a)). Art. 63, page 8 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

(12) Lexical stress in Catalan a. [-C#] b. [-V#] c. Antepenultimate stress [kəˈpitul] [təˈβu] [məˈtɾɔpuli] ‘chapter’ ‘tabu’ ‘metropolis’ [ˈʎapis] [∫impənˈze] [ˈpaʒinə] ‘pencil’ ‘chimpanzee’ ‘page’ [nuˈmεɾik] [kəˈfε] [ˈaβitət] ‘numerical’ ‘coffee’ ‘habitat’ [təˈlεfun] [pəɾuˈne] [ˈlawðənum] ‘telephone’ ‘fibula’ ‘laudanum’

Most accounts of Catalan stress are couched within moraic theory and assume a right- aligned moraic trochee (i.e. left-headed foot) to be the unmarked foot in the language (Cabré 1993; Cabré & Kenstowicz 1995; Serra 1996; Vallverdú 1997; Bonet & Lloret 1998; Bonet & Torres-Tamarit 2010). These studies would agree on the algorithm in (13).

(13) Catalan stress algorithm a. Project a mora from every vowel In the absence of lexical stress: b. Project a mora from a final consonant c. Build a right-aligned moraic trochee If a vowel is stressed lexically: d. Build that foot with its strong branch on the lexically stressed vowel.

Some representations appear in (14). In (14a), the word ends in a consonant, so a mora is projected and the right-aligned trochaic foot is built. In (14b), the right-aligned trochaic foot is built similarly onto two vowels. In (14c), the word has lexical stress on the first vowel, marked with an accent, so the final mora is ignored and the foot is built on top of the lexically stressed vowel.

(14)(14) CatalanCatalan stress stress with with moras moras and feetand feet a. F b. F c. F µ µ µ µ µ µ µ µ µ

p @ t i t k @ n a R i p a´ Z i n @

This account is not easily extended to cases of final /n/ deletion. One needs to assume that final This/n/, likeaccount all other is not final easily consonants, extended projects to cases a mora of evenfinal when /n/ itdeletion. is not realized. One needs An alternative to assume that finalwould be/n/, to like assume all that,other even final though consonants, the final /n/ projectsis not realized,a mora itseven position when projectsit is not a mora.realized. An alternativeHowever, combining would be skeletal to assume positions that, with even moras though is quite the unorthodox, final /n/ as is morasnot realized, were precisely its position projectsposited to a get mora. rid of However, skeletal slots combining (among other skeletal goals). positions Finally, thiswith account moras has is nothingquite unorthodox, to say asabout moras the relationwere precisely between finalposited/n/ deletionto get rid and of stress, skeletal which slots we saw(among was systematic:other goals)./n/ Finally,is thisnot deleted account when has the nothing preceding to vowelsay about is unstressed the relation (though between see footnote final 2). /n/ deletion and stress, whichBut we even saw independently was systematic: of /n/ deletion,/n/ is not the deleted moraic accountwhen the faces preceding a conceptual vowel problem. is unstressed Its (thoughreasoning see is circular: footnote the 2). moraic status of final consonants is used to explain final stress, which is also the only reason to accord final consonants moraic status. Therefore, if an account of But even independently of /n/ deletion, the moraic account faces a conceptual problem. Catalan stress can be provided that relies not on moras but rather on some other independently- Its reasoning is circular: the moraic status of final consonants is used to explain final motivated aspect of phonology, it would be preferable to a moraic account. Working within stress,Strict CV, which Scheer is also and the Szigetv onlyari´ reason (2005) to show accord that final empty consonants nuclei, an independently moraic status. motivated Therefore, if anaspect account of their of theory,Catalan can stress replace can moras. be provided Moras, theythat say, relies can not be gottenon moras rid of. but We rather will now on some see that this is true for the Catalan system. Moreover, such an account of Catalan stress paves the way to a better understanding of final /n/ deletion.9

3 Analysis

3.1 Stress with empty nuclei rather than moras We ended the preceding section by stating the need for something that would explain the special status of final consonants. In the section on GP, we saw how that theory provides exactly this something: final consonants are a special kind of onset. They are followed by an empty nucleus.The theory therefore provides the generalization for Catalan regular stress in (15).

(15) Catalan stress algorithm in GP In the absence of lexical stress: a. Build a right-aligned trochaic foot based on nuclei, empty or full; only full nuclei can be foot heads If a vowel is stressed lexically: b. Build that foot with its strong branch on the lexically-stressed vowel.

The results of the algorithm for stress assignment in Catalan are represented in (16). In (16a), a right-aligned foot is built whose dependent is the FEN. In (16b), the dependent is now a full vowel. This way, final and penultimate stress are derived with the same type of foot. In (16c), the foot is not right-aligned because the first vowel is lexically stressed.

9See Ulfsbjorninn (2014) for an elaboration on Scheer and Szigetvari’s´ (2005) model.

9 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 9 of 23

other independently-motivated aspect of phonology, it would be preferable to a moraic account. Working within Strict CV, Scheer and Szigetvári (2005) show that empty nuclei, an independently motivated aspect of their theory, can replace moras. Moras, they say, can be gotten rid of. We will now see that this is true for the Catalan system. Moreover, such an account of Catalan stress paves the way to a better understanding of final /n/ deletion.9 3 Analysis 3.1 Stress with empty nuclei rather than moras We ended the preceding section by stating the need for something that would explain the special status of final consonants. In the section on GP, we saw how that theory provides exactly this something: final consonants are a special kind of onset. They are followed by an empty nucleus. The theory therefore provides the generalization for Catalan regular stress in (15).

(15) Catalan stress algorithm in GP In the absence of lexical stress: a. Build a right-aligned trochaic foot based on nuclei, empty or full; only full nuclei can be foot heads If a vowel is stressed lexically: b. Build that foot with its strong branch on the lexically-stressed vowel.

The results of the algorithm for stress assignment in Catalan are represented in (16). In (16a), a right-aligned foot is built whose dependent is the FEN. In (16b), the dependent is now a full vowel. This way, final and penultimate stress are derived with the same type of foot. In (16c), the foot is not right-aligned because the first vowel is lexically stressed.

(16)(16) CatalanCatalan stress stress with with empty empty nuclei nuclei and no and moras no moras a. F b. F c. F

(16) CatalanC stressV C withV emptyC V nuclei andC V no morasC V C V C V C V C V a. F b. F c. F p @ t i t k @ n a R i p a´ Z i n @ C V C V C V C V C V C V C V C V C V This proposal does not suffer from the circularity of the mora-based view because FEN are an independentlyThis proposalp motivated does@ t noti phonologicalsuffert from tool.thek @circularity Then usea of Rof FEN’s thei inmora-based stressp a´ assignmentZ viewi nbecause is@ not ad FEN are hocan independentlyand therefore not motivated problematic phonological methodologically. tool. The use of FEN’s in stress assignment is ThisnotFurthermore, proposalad hoc and does therefore since not suffer stress not fromis now problematic the tied circularity not to finalmethodologically. of consonantsthe mora-based but rather view because to the FEN, FEN the are non- an independentlyrealizationFurthermore, of the motivated /n/sinceis notstress phonological as problematicis now tool.tied as not The in the to use moraicfinal of FEN’s account.consonants in stress To anticipate assignmentbut rather our to is analysis, notthe adFEN, the hocconsidernon-realizationand therefore the representation notof problematicthe /n/ in (17),is not methodologically. which as problematic depicts a final as/n/ in theas floating moraic below account. its C position.To anticipate SinceourFurthermore, analysis, the CV structure consider since stress is stillthe is there, nowrepresentation tiedthe stress not to algorithm final in consonants(17), is perfectlywhich but depicts rather able to to takea the final FEN,it into /n/ the account. non-as floating realization of the /n/ is not as problematic as in the moraic account. To anticipate our analysis, Webelow will its explain C position. the generation Since the of the CV skeleton structure and is the still floating there,/n/ thein thestress next algorithm section. is perfectly consider the representation in (17), which depicts a final /n/ as floating below its C position. able to take it into account. We will explain the generation of the skeleton and the floating Since(17) theStress CV structure with unrealized is still there, final the/n/ stressand FEN algorithm is perfectly able to take it into account. We/n/ will in the explain next the section. generationF of the skeleton and the floating /n/ in the next section.

(17)(17) StressStressC V with withC unrealized VunrealizedC V final final/n/ and /n/ FEN and FEN F k u m u n C V C V C V

k u m u n

The FEN-based, mora-less account in Strict CV is superior to the mora-based account because it 9 reliesSee Ulfsbjorninn not on the (2014) realization for an of elaboration/n/ but just on onScheer the presenceand Sziget ofvári’s the (2005) CV structure model. that is related to the final /n/. Of course, the structure in (17) is quite abstract, in the sense that there is an Theentire FEN-based, “syllable” mora-less that is not account associated in Strictto any CV segmental is superior material. to the This mora-based fact, as weaccount will see because in the itnext relies subsection, not on the is realization the key to of understanding/n/ but just on the the interaction presence of of the the process CV structure of final that/n/ isdeletion related towith the stress. final /n/ Note. Of that course, from now the structure on, we use inthe (17) term is quite deletion abstract, descriptively: in the sense we that assume there that is thean entireunderlying “syllable”/n/ is that not is deleted, not associated but floats to any below segmental its C position. material. This fact, as we will see in the next subsection, is the key to understanding the interaction of the process of final /n/ deletion with stress. Note that from now on, we use the term deletion descriptively: we assume that the 3.2 Final /n/ deletion underlying /n/ is not deleted, but floats below its C position. We claim that /n/ in Catalan is placeless. This is why it undergoes place assimilation to a 3.2following Final /n/ consonant.deletion In such circumstances, /n/ is, in the terms used by Scheer (2004), a kind of “partial” or “nasal geminate” that is protected from deletion. The nasal is licensed because Wethe claim place nodethat /n/ of thein Catalan following is consonant placeless. branches This is why and it links undergoes to the position place assimilation occupied by to the a followingnasal, aspreviously consonant. shown In such in circumstances, (9). In a non-final/n/ is, onset in the position terms used followed by Scheer by a full(2004), vowel, a kind the ofnasal “partial” is also or licensed—recall “nasal geminate” that that all is onsets protected are licensed from deletion. by an immediately The nasal is following licensed because realized thenucleus. place node This leavesof the followingone position consonant where /n/ branchesis not andlicensed: links tobefore the position the FEN. occupied This is exactly by the nasal,the position as previously where final shown/n/ inis (9). usually In a deleted.non-final We onset may position now understand followed by why a full final vowel,/n/ is the not nasalrealized: is also as statedlicensed—recall in the generalization that all onsets in (18), are licensed/n/ in Catalan by an immediately must be licensed following by a following realized nucleus.segment, This either leaves a full one vowel position or a following where /n/ onsetis not from licensed: which to before receive the place FEN. of This articulation. is exactly the position where final /n/ is usually deleted. We may now understand why final /n/ is not realized:(18) Catalan as stated/n/ in the generalization in (18), /n/ in Catalan must be licensed by a following segment,In either Catalan, a full/n/ vowelmust or be a followinglicensed by onset a following from which segment. to receive Otherwise, place of it floats.articulation.

(18) Catalan /n/ In Catalan, /n/ must be licensed by a following segment. Otherwise, it floats. 10

10 Art. 63, page 10 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

The FEN-based, mora-less account in Strict CV is superior to the mora-based account because it relies not on the realization of /n/ but just on the presence of the CV structure that is related to the final /n/. Of course, the structure in (17) is quite abstract, in the sense that there is an entire “syllable” that is not associated to any segmental material. This fact, as we will see in the next subsection, is the key to understanding the interaction of the process of final /n/ deletion with stress. Note that from now on, we use the term deletion descriptively: we assume that the underlying /n/ is not deleted, but floats below its C position. 3.2 Final /n/ deletion We claim that /n/ in Catalan is placeless. This is why it undergoes place assimilation to a following consonant. In such circumstances, /n/ is, in the terms used by Scheer (2004), a kind of “partial” or “nasal geminate” that is protected from deletion. The nasal is licensed because the place node of the following consonant branches and links to the position occupied by the nasal, as previously shown in (9). In a non-final onset position followed by a full vowel, the nasal is also licensed—recall that all onsets are licensed by an immediately following realized nucleus. This leaves one position where /n/ is not licensed: before the FEN. This is exactly the position where final /n/ is usually deleted. We may now understand why final /n/ is not realized: as stated in the generalization in (18), /n/ in Catalan must be licensed by a following segment, either a full vowel or a following onset from which to receive place of articulation.

(18) Catalan /n/ In Catalan, /n/ must be licensed by a following segment. Otherwise, it floats.

A look at (17) above shows that no such licensing is available when /n/ is the last consonant of the word.10 If so, we have a hypothesis regarding why /n/ remains afloat word-finally. Still, we have seen that floating is possible only when the preceding vowel is stressed. Ifitis not, for some reason we do find final [n]’s (e.g. origen [uˈɾiʒən]). This implies that our generalization regarding Catalan /n/ in (18) is not always met, that is, it is sometimes violated, and violability is best dealt with within the framework of OT. Therefore, the OT constraint in (19a) is proposed. The second constraint (19b) concerns the justification of CV units. As for any other piece of representation, we assume that if there is no way to engage a CV unit, its inclusion in the representation is dispreferred. We term this constraint *Unengaged-CV;11 by “engaged” we mean “associated to material from another level of representation”.

(19) Constraints a. *Unlicensed[n] An /n/ linked to its C unit must be licensed by a following segment. b. *Unengaged-CV Every CV unit must be engaged by some other level of representation (either segmental or metrical).

10 An interpretation along these lines was also suggested in Jiménez and Lloret (2011), according to which coronal nasals, which are typologically less marked and perceptually weaker than labial or dorsal nasals, can be deleted at a lesser cost. This idea of preservation of the marked was previously considered in Bonet et al. (2004, 2005) and Wheeler (2005) (see section 3.3.4 for more details). 11 *Unengaged-CV could also be named Parse-CV. This constraint expresses a widespread idea in phonological theory that phonological structure must be parsed at appropriate levels of representation. Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 11 of 23

Before we present our analysis, it is crucial to make our assumptions about the generation of CV units explicit. We assume a containment theory of faithfulness (Prince & Smolensky 1993/2004) by which no segmental material present in the input can be deleted in the output.12 Given the fact that the Strict CV structure is universal, it is an unviolable property of the GEN function of the grammar to project a uniform CV structure from any segmental string. For instance, an input form ending in a consonant will project in all languages a CV unit. What GEN can manipulate is the specific association of the segmental string with the universally-projected CV structure. As a consequence of this, GEN is able to leave segments afloat, or add association lines. The set of candidates generated is evaluated by the function EVAL according to the language-particular hierarchy of universal constraints as in standard OT. Now consider the tableau in (20), with an input containing a final /n/. According to the general algorithm of Catalan stress in (15), the FEN is engaged by the foot and parsed as its dependent. The constraint *Unlicensed[n] is not violated by candidate (a), the winning candidate, because final /n/ remains afloat. Associating the final /n/ is unwarranted because it would be unlicensed, as [n] would not be followed by any segment (candidate b), and inserting a vowel to occupy the final V position violates Dep-V, the anti-insertion constraint (candidate c).13 This evaluation demonstrates that both Dep-V and *Unlicensed[n] dominate *Floating, a constraint against floating segments.ING, a constraint In sum, against the foot floating parses segments. the last In sum,CV unit the footand parses this is the precisely last CV unit what and allows this is /n/ deletionprecisely in what a position allows /n/ in deletionwhich it in would a position otherwise in which be it wouldunlicense otherwised. be unlicensed. (20)(20) TableauTableau forfor[ku"mu] [ku’mu] /kumun/ DEP-V *UNENGAGED-CV *UNLICENSED[n] *FLOATING F

C V C V C V

a. k u m u n * F

C V C V C V

b. k u m u n *! F

C V C V C V

c. k u m u n @ *! If so, representing the right-aligned foot with a FEN dependent, allows one not only to get rid Ifof so, moras representing but also to the leave right-aligned afloat the weak foot final with/n/ a. FEN We will dependent, now see that allows when one the not FEN only is not to get ridengaged of moras by metrical but also structure, to leave/n/ afloatcannot the remain weak afloat. final /n/. We will now see that when the FEN is not engaged by metrical structure, /n/ cannot remain afloat. 3.3 Instances of word-final [n] 3.3.1 Blocking of final /n/ deletion before unstressed vowels 12The The situationchoice of incontainment words with is finalmotivated/n/ that by the have Catalan lexical, facts: non-final words stresswith final is different. deleted /n/ Now are con- regularly siderstressed the like tableau C-final in (21) words, for the which word ["kOlun] require ‘colon’. the nasal In thisto case,be contained if /n/ does in not the associate output to even though itis unrealized. 13its One C reviewer position argues and the that final assuming CV unit a isviolation not engaged of Dep-V by anyby epenthetic level of representation vowels is incompatible because thewith the footbasic is management not right-aligned, of empty thus nuclei violating in Strict *U CV,NENGAGED whereby-CV realizing (candidate ungoverned a). The empty final V’s/n/ is mustnot costly. associateHowever, tothis its is C not position precise. in In order this case to engage we are the dealing otherwise with a unengaged final empty CV nucleus. unit, and FEN’s the in can- Strict CV are parametrically licensed. In other words, like governed empty nuclei, they should remain unrealized. didateTherefore, violating realizing *U themNLICENSED does come[n] withwins a cost out. under This any evaluation view. Let us demonstrates also clarify that that we both use the DEP constraint-V andname *U DepNENGAGED, characteristic-CV of dominate Correspondence *UNLICENSED Theory, instead[n]. The of constraintFill, although FAITHFULNESS we assume containment-STRESS (see mustthe paragraph also dominate after (19)). *UNLICENSED[n] in order to ensure adherence to lexical stress (which appears marked as a standard IPA stress diacritic in the underlying representation).14

14One reviewer wonders how lexical stress can be encoded in our model if CV units are absent from underlying representations. We assume that lexical stress can be modeled by associating the relevant vowel with a V slot which is a foot’s head in the lexical representation, rather than through the algorithm (that is, by assuming underlying prosodic structure).

12 Art. 63, page 12 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

3.3 Instances of word-final [n] 3.3.1 Blocking of final /n/ deletion before unstressed vowels The situation in words with final /n/ that have lexical, non-final stress is different. Consider the tableau in (21) for the word [‘kɔlun] ‘colon’. In this case, if /n/ does not associate to its C position and the final CV unit is not engaged by any level of representation because the foot is not right-aligned, thus violating *Unengaged-CV (candidate a). The final /n/ must associate to its C position in order to engage the otherwise unengaged CV unit, and the candidate violating *Unlicensed[n] wins out. This evaluation demonstrates that both Dep-V and *Unengaged-CV dominate *Unlicensed[n]. The constraint Faithfulness-stress must also dominate *Unlicensed[n] in order to ensure adherence to lexical stress (which appears marked as a standard IPA stress diacritic in the underlying representation).14

(21)(21) TableauTableau for for["kOlun] [ˈkɔlun] /"kOlun/ FAITH-STRESS DEP-V *UNENGAGED-CV *UNLICENSED[n] *FLOATING F

C V C V C V

a. k O l u n *! * F

C V C V C V

b. k O l u n * F

C V C V C V

c. k O l u n @ *! F

C V C V C V

d. k u l u n *! *

The ranking of DEP-V above *UNLICENSED[n] ensures that unlicensed [n]’s are not repaired Thethrough ranking realization of Dep-V of above the FEN *Unlicensed as a full vowel[n] through ensures vowel that unlicensed epenthesis, which[n]’s are is schwanot repaired in throughCatalan (comparerealization candidates of the FEN b and as c ina full tableau vowel 20). through The ranking vowel of *Uepenthesis,NENGAGED which-CV above is schwa in*U CatalanNLICENSED (compare[n] expresses candidates the fact b and that c [n] in can tableau be unlicensed 20). The only ranking in those of cases*Unengaged-CV in which abovethe metrical *Unlicensed structure[n] cannot expresses engage athe CV fact unit that due [n] to a can lexical be stressunlicensed that pushes only thein those foot far cases infrom which the the right metrical word edge structure (compare cannot candidates engage a and a CV b in unit tableau due 21).to a Finally,lexical stress the ranking that pushes of *UNLICENSED above *FLOATING derives the fact that when the last CV unit is engaged the foot far from[n] the right word edge (compare candidates a and b in tableau 21). Finally, by the metrical structure, a floating segment is preferred over an unlicensed, associated [n] the ranking of *Unlicensed[n] above *Floating derives the fact that when the last (compare candidates a and b in tableau 20). Thus, our analysis captures not only final /n/ CVdeletion, unit is but engaged also its non-application by the metrical when structure, the preceding a floating vowel is not segment stressed. is preferred over an unlicensed, associated [n] (compare candidates a and b in tableau 20). Thus, our analysis captures not only final /n/ deletion, but also its non-application when the preceding 3.3.2 Final /n/ deletion counterfed by final consonant cluster simplification vowel is not stressed. Catalan displays a process of consonant cluster simplification in final position by which a word- 3.3.2final Final stop deletes /n/ deletion when preceded counterfed by a homorganic by final consonant nasal (22). Wecluster therefore simplification assume that under- Catalanlyingly thedisplays words a in process (23a) are of/dent/ consonant, /kanp/ clusterand /sang/ simplification. in final position by which a word-final stop deletes when preceded by a homorganic nasal (22). We therefore assume (22) Final /nC/ cluster simplification in Catalan that underlyinglya. Underived the wordsb. Derivedin (23a) are /dent/, /kanp/ and /sang/. ["den] [d@n"tal] 14 One reviewer wonders‘tooth’ how lexical‘dental’ stress can be encoded in our model if CV units are absent from underlying representations. We assume that lexical stress can be modeled by associating the relevant vowel with a V slot which is a foot’s head in the lexical representation, rather than through the algorithm (that is, by assuming underlying["kam] prosodic structure).[k@mp@"ROl] ‘field’ ‘peasant’

["saN] [s@Ngu"nos] ‘blood’ ‘bloody’

A word like ["den] illustrates the fact that [n] surfaces word-finally when it has come to stand in this position due to cluster simplification. Although the underlying /t/ in /dent/ is not realized—it remains afloat below its C position—we claim that it has the ability to license the

13 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 13 of 23

(22) Final /nC/ cluster simplification in Catalan a. Underived b. Derived [ˈden] [dənˈtal] ‘tooth’ ‘dental’ [ˈkam] [kəmpəˈɾɔl] ‘field’ ‘peasant’ [ˈsaŋ] [səŋguˈnos] ‘blood’ ‘bloody’

A word like [ˈden] illustrates the fact that [n] surfaces word-finally when it has come to stand in this position due to cluster simplification. Although the underlying /t/ in /dent/ is not realized—it remains afloat below its C position—we claim that it has the ability to license the preceding nasal. This is evident from the fact that the nasal acquires the place of articulation of the following non-realized stop. precedingFirst, consider nasal. Thisthe isexample evident from[ˈsaŋ]. the factThe thatnasal the surfaces nasal acquires as velar the place because of articulation it acquires of the velarthe following place of non-realized articulation stop. of the following underlying velar stop, and therefore the non- realizedFirst, stop consider licenses the example the preceding["saN]. The nasal. nasal We surfaces propose, as velar following because itcontainment, acquires the velar that the velarplace stop of articulation is not deleted of the in following the output, underlying but is velaronly stop,delinked and therefore from its the C position. non-realized This stop allows forlicenses a branching the preceding structure nasal. at Wethe propose,C-place followingnode (Clements containment, & Hume that 1995) the velar originally stop is not linked todeleted the final in the stop. output, This but C-place is only delinkednode branches from its and C position.links to the This V allows position for aoccupied branching by the nasal,structure as in at the(23). C-place node (Clements & Hume 1995) originally linked to the final stop. This C-place node branches and links to the V position occupied by the nasal, as in (23).

(23)(23) Nasal-obstruent cluster cluster in ["saN]in [ˈsaŋ] C V C V C V

s a N g

C-place

[dor]

The same structure as in (23) can be attributed to [ˈden]-type cases. We assume that /t/, like /n/, is underspecified for place because it also undergoes place assimilation in pre-consonantalThe same structure position as in (23) ([ˈs canεk be ˈkazəs] attributed ‘seven to ["den] houses’-type cf. cases. [ˈsεk] We ‘seven’). assume that However,/t/, like it is a/n/ fact, is of underspecified Catalan that, for unlike place because /n/, /t/ it alsodoes undergoes not need place to be assimilation licensed inin pre-consonantalfinal position. We formalizeposition ( ["sEkthis "kaz@s]by representing‘seven houses’ /t/, cf. but["sEt] not ‘seven’ /n/, as). However,having an it isunspecified a fact of Catalan C-place that, node, thatunlike is, /n/with, /t/ nodoes terminal not need feature. to be licensed The nasal in final in [‘den] position. can We be formalize associated this byto its representing C position in order/t/, but to notsatisfy/n/, as*Floating having an unspecifiedwithout violating C-place node,*Unlicensed that is, with[n] no because terminal the feature. /t/ Thestill has nasal in ["den] can be associated to its C position in order to satisfy *FLOATING without vio- anasal C-place in ["den] node,can which be associated allows for to its the C positionsame branching in order to structure satisfy *F foundLOATING in [ˈsaŋ]without (24). vio- The lating *UNLICENSED[n] because the /t/ still has a C-place node, which allows for the same onlylating difference *UNLICENSED with[n] respectbecause to the[ˈsaŋ]/t/ stillis that has in a C-place[ˈden], node,the underlying which allows /t/ for has the no same terminal branching structure found in ["saN] (24). The only difference with respect to ["saN] is that in placebranching feature. structure found in ["saN] (24). The only difference with respect to ["saN] is that in ["den], the underlying /t/ has no terminal place feature. (24) Representation for ["den] (24)(24) Representation for for["den] [ˈden] C V C V C V

d e n t

C-place

The fact that /t/ is not linked to C and is therefore left unpronounced certainly violates *UNENGAGED- CV. However, this violation is motivated because *UNENGAGED-CV is dominated by an inde- pendent markedness constraint against homorganic nasal-stop clusters in word-final position, which we refer to as *NT#, as seen in the tableau in (25). Both candidates (a) and (b) sat- isfy *UNLICENSED[n]: candidate (a) because of the C-place branching node, and candidate (b) because the nasal is unassociated. Candidate (b), with non-realization of /n/, violates

14 Art. 63, page 14 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

The fact that /t/ is not linked to C and is therefore left unpronounced certainly violates *Unengaged-CV. However, this violation is motivated because *Unengaged-CV is dominated by an independent markedness constraint against homorganic nasal-stop clusters in word-final position, which we refer to as *NT#, as seen in the tableau in (25). Both candidates (a) and (b) satisfy *Unlicensed[n]: candidate (a) because of the C-place branching node, and candidate (b) because the nasal is unassociated. Candidate (b), with non-realization of /n/, violates *Floating twice, and this is why candidate (a) wins. Foot *FLOATING twice, and this is why candidate (a) wins. Foot boundaries are marked in (25) with boundariesparentheses. are marked in (25) with parentheses.

(25)(25) TableauTableau forfor["den] [ˈden] /dent/ *NT# *UNENGAGED-CV *UNLICENSED[n] *FLOATING (C V C V) C V

d e n t

C-place

a. * * (C V C V) C V

d e n t

C-place

b. * **! (C V C V) C V

d e n t

C-place

c. *!

One reviewer raises two relevant empirical facts about ["den]-type cases. First, when clus- Oneter simplification reviewer raises counterfeeds two relevant final /n/ empiricaldeletion, thefacts nasal about is alveolar, [‘den]-type as are cases. all final First,[n]’s in when clusterCatalan. simplification However, when counterfeeds coronal nasals final precede /n/ adeletion, realized [t]theor nasal[d], whichis alveolar, are always as are den- all final [n]’stal in in Catalan, Catalan. the However, nasal is also when dental coronal ([d@n”"t”al] nasals). If precede/t/ is assumed a realized to be [t] only or specified[d], which for are alwaysa C-place dental node in without Catalan, a terminal the nasal feature, is also and dental final [n] [dəisn̪ licensedˈt̪al]. If by/t/ an is associationassumed to to be that only specifiedC-place node for botha C-place in ["den] nodeand without in [d@n”"t”al] a terminal, why is thefeature, final [n]andin final["den] [n]not is dental? licensed We by an associationclaim that onlyto thatrealized C-place/t/ (and node/d/ both) is dentalin [ˈden] in Catalan. and in This[dən̪ claimˈt̪al], canwhy be is expressed the final as a[n] in [ˈden]markedness not dental? implicational We claim constraint that ofonly the realized type “if realized/t/ (and[t] /d/), then is [dental]” dental (seein Catalan. Levelt & This claimvan Oostendorp can be expressed 2007, van as Oostendorp a markedness 2014 implicational and Torres-Tamarit constraint et al. 2017 of the for type implicational “if realized [t],constraints then [dental]” of the if-then-type). (see Levelt Such& van output-oriented Oostendorp 2007, constraints van Oostendorp are independently 2014 needed and Torres- to Tamaritaccount et for al. how 2017 certain for implicational classes of segments, constraints such as of rhotics, the if-then-type). are interpreted Such by theoutput-oriented phonetics on a language-particular basis. This constraint would not affect unrealized /t/’s, which would constraints are independently needed to account for how certain classes of segments, explain the alveolar realization of ["den]. The second empirical fact is the following: when a such as rhotics, are interpreted by the phonetics on a language-particular basis. This word like ["den] is followed by a consonant-initial word, the /n/ acquires the place of articula- constrainttion of that would following not consonant,affect unrealized as in ["dem /t/’s, p@"tit@] which‘small would tooth’ explain (cf.*["den the p@"tit@] alveolar). How realization can of/n/ [ˈden].assimilate The second to a following empirical labial fact through is the an following: intervening when/t/? Wea word consider like this [ˈden] to be is a followed case byof a indirect consonant-initial place assimilation. word, Thethe /n//t/ isacquires present inthe the place representation, of articulation and its of C-place that following node consonant,licenses the as previous in [ˈdem nasal. pəˈtitə] The fact‘small that tooth’ the C-place (cf. *[ˈden node associated pəˈtitə]). to How the /t/ canis /n/ empty assimilate trig- togers a following a place-sharing labial configuration through an betweenintervening the feature /t/? We associated consider to thethis following to be a case word-initial of indirect placeconsonant assimilation. and the C-place The /t/ node is present of the /t/ in (recallthe representation, that this also happens and its whenC-place/t/ isnode realized licenses theand previous precedes nasal. a consonant, The fact as that in set the cases C-place["sEk "kaz@s]node associated‘seven houses’). to the Because/t/ is empty the nasal triggers is a place-sharinglinked to the C-placeconfiguration node of the between/t/ for independentthe feature reasons,associated the nasalto the indirectly following inherits word-initial the consonantplace specification and the of C-place the following node word-initial of the /t/ consonant, (recall that as illustrated this also in happens (26). This when analysis /t/ is realizedis non-stratal and precedes (cf. the analysis a consonant, of Catalan as in in set Kiparsky cases [ˈs 1985,εk ˈkazəs] in which ‘seven dentalization houses’). only Because takes the nasalplace is postlexically), linked to the and C-place does not node violate of the containment. /t/ for independent reasons, the nasal indirectly

15 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 15 of 23

inherits the place specification of the following word-initial consonant, as illustrated in (26).(26) ThisNasal analysis place is assimilation non-stratal across(cf. the word-boundaries analysis of Catalan in Kiparsky 1985, in which dentalizationC onlyV takesC placeV postlexically),C and does not violate containment.

(26)(26) Nasal... n placeplace assimilation assimilationt across across## word-boundaries word-boundariesp ... C V C V C C-place C-place ... n t ## p ...

C-place C-place[labial]

[labial]

TheThe absence absence of of final final /n//n/ deletiondeletion inin pluralsplurals asas inin ["pans][ˈpans]‘breads’ ‘breads’ (cf.(cf. ["pa][ˈpa] ‘bread’)‘bread’) receivesreceives the thesame same explanation explanation as ["den] as . [ˈden]. The only The difference only difference is that in ["pans] is that, as opposedin [ˈpans], to ["den] as, theopposed licen- to The absence of final /n/ deletion in plurals as in ["pans] ‘breads’ (cf. ["pa] ‘bread’) receives the [ˈden],sor [s] thesurfaces licensor because [s] surfaces *NT# becauseexclusively *NT# refers exclusively to homorganic refers to nasal-stop homorganic clusters. nasal-stop An eval- same explanation as ["den]. The only difference is that in ["pans], as opposed to ["den], the licen- clusters.uation of An the evaluation form of isthe given form in [ˈpans] (27). Note is given that in ,(27). the result Note ofthat applying [s], the final-obstruent result of sor [s] surfaces because["pans] *NT# exclusively refers to homorganic[s] nasal-stop clusters. An eval- applyingdevoicing final-obstruent to the plural morph devoicing/z/, isto specifiedthe plural for morph [coronal]. /z/, is Wespecified assume for this [coronal]. representation We assumeuation ofthis the representation form ["pans] is givenbecause in (27). /z/ is Note a perfect that [s] ,pre-consonantal the result of applying “coda” final-obstruent in Catalan and becausedevoicing/z/ tois the a plural perfect morph pre-consonantal, is specified “coda” for [coronal]. in Catalan We and assume does not this undergo representation place assimi- does not undergo place assimilation./z/ The result is a licensed nasal specified for [coronal], lation.because The/z/ resultis a perfect is a licensed pre-consonantal nasal specified “coda” in for Catalan [coronal], and does a redundant not undergo specification place assimi- for nasals a redundant specification for nasals in Catalan. inlation. Catalan. The result is a licensed nasal specified for [coronal], a redundant specification for nasals in Catalan. (27)(27) Tableau for ["pans][ˈpans] (27) Tableau/pan+z/ for ["pans] *NT# *UNENGAGED-CV *UNLICENSED[n] *FLOATING *NT# *UNENGAGED-CV *UNLICENSED[n] *FLOATING /pan+z/(C V C V) C V (C V C V) C V p a n s p a n s

C-placeC-place

a.a. [cor][cor] (C(C VV C C V)V) C C V V

pp a a n n s s

C-placeC-place

b. [cor] *! b. [cor] *! (C V C V) C V (C V C V) C V p a n s p a n s C-place C-place c. [cor] *! ** c. [cor] *! ** With the addition of the constraint *NT#, we have easily extended our account to another case Withwhere the final addition/n/ deletion of the is constraint blocked. *NT#, we have easily extended our account to another case Withwhere the final addition/n/ deletion of the constraint is blocked. *NT#, we have easily extended our account to another case3.3.3 where No final final/n/ deletion/n/ deletion in final is /rn/blocked.clusters A distinct case of realization of word-final is represented by the cases in (4a), like 3.3.33.3.3 No No final final /n//n/ deletiondeletion in infinal final /rn//rn/ clusters/n/clusters ["karn] “meat”. Word-finally, /n/ surfaces if it is preceded by [r], the most sonorous among sonorant AAconsonants. distinct distinct case 15 A of theory realizationrealization in which ofof word-final/n/word-finalis a coda/n/ /n/ inis both is represented represented/kumun/ byand theby/karn/ the cases caseswill in (4a),runin (4a), into like like["karn] [ˈkarn] “meat”. Word-finally, /n/ surfaces if it is preceded by [r], the most sonorous among “meat”.15 Word-finally, /n/ surfaces if it is preceded by [r], the most sonorous among sonorant Final [n] can15 also marginally surface before the glide [j]. The only Catalan words with a word-final [-jn] consonants.cluster, however, areA the theory following in three which technical/n/ loanwords:is a codaain ["ajn] in both, or ["ain]/kumun/, zain ["zajn]and, or ["zain]/karn/, Aramaicwill run into 15Final [n] can also marginally surface before the glide [j]. The only Catalan words with a word-final [-jn] cluster, however, are the following three technical loanwords:16 ain ["ajn], or ["ain], zain ["zajn], or ["zain], Aramaic

16 Art. 63, page 16 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

sonorant consonants.15 A theory in which /n/ is a coda in both /kumun/ and /karn/ will run into difficulties in explaining deletion in the former yet retention in the latter. Strict CV, in contrast, offers an independently motivated solution. Although appearing in final position,difficulties the in segment explaining preceding deletion in /n/ the formerin a Strict yet retention CV representation in the latter. Strictis not CV, a stressed in contrast, vowel butoffers rather an independently an internal empty motivated nucleus. solution. As Althoughshown in appearing (28), because in final feet position, can only the segment be headed bypreceding full vowels,/n/ in thisa Strict empty CV representation nucleus cannot is not head a stressed a right-aligned vowel but rather foot an with internal a dependent empty FEN.nucleus. Instead, As shown the inV (28),associated because to feet /a/ can heads only bethe headed foot. byThe full CV vowels, unit thisabove empty the nucleus /n/ is not engagedcannot head in the a right-aligned metrical structure. foot with aThis dependent configuration FEN. Instead, is parallel the V associated to the toone/a/ foundheads when /n/the precedes foot. The CVan unitunstressed above the vowel/n/ is due not engagedto lexical in thestress metrical seen structure.above in This (2). configuration Thus, the same is parallel to the one found when precedes an unstressed vowel due to lexical stress seen mechanism that we propose to explain/n/ the blocking of final /n/ deletion before unstressed above in (2). Thus, the same mechanism that we propose to explain the blocking of final /n/ vowels explains the [ˈkarn]-type cases. This is a clear advantage of our analysis. As can be deletion before unstressed vowels explains the ["karn]-type cases. This is a clear advantage of seen in tableau (28), the need to satisfy the constraint *Unengaged-CV, which dominates our analysis. As can be seen in tableau (28), the need to satisfy the constraint *UNENGAGED- *UnlicensedCV, which dominates[n], triggers *UNLICENSED the realization[n], triggers of final the realization /n/. of final /n/.

(28)(28) Tableau for for["karn] [ˈkarn] /karn/ DEP-V *UNENGAGED-CV *UNLICENSED[n] *FLOATING F

C V C V C V

a. k a r n *! * F

C V C V C V

b. k a r n * F

C V C V C V

c. k a r n @ *!

This concludes the account of the phonological deletion of coronal nasals in word-final posi- Thistion. concludes We can now the move account to the retentionof the phonological of other related deletion segments. of coronal nasals in word-final position. We can now move to the retention of other related segments.

3.3.43.3.4 A A note note on on the the non-deletion non-deletion of of non-coronal non-coronal nasals nasals and and non-nasal non-nasal coronals OneOne important important questionquestion is whyis why/n/ /n/deletes deletes in word-final in word-final position whereasposition non-coronalwhereas non-coronal nasals nasalslike labial like /m/labialor palatal/m/ or/ñ/ palataldo not. /ɲ/ Unlike do not./n/, theseUnlike nasals /n/, are these specified nasals for are place. specified16 This for place.simple16 factThis makes simple them fact automatically makes them licensed. automatically In OT terms, licensed. this could In beOT seen terms, as an this effect could of be seenpreservation as an effect of place of features.preservation of place features. However,However, there there are otherother non-nasal non-nasal coronals coronals that are that never are deleted. never Fordeleted. example, For the example, coronal the coronalobstruents obstruents/t, s/ and the /t, non-nasal s/ and the coronal non-nasal sonorant coronal/l/ are never sonorant deleted /l/ in word-final are never position. deleted in We already discussed in 3.3.2 and showed that although it is underspecified it does have a word-final position./t/ We already discussed /t/ in 3.3.2 and showed that although itis C-place node, and therefore it does not require licensing and is not expected to delete in final underspecified it does have a C-place node, and therefore it does not require licensing position. However, unlike /t/, the coronal fricative [s] and the non-nasal coronal sonorant [l] andnever is assimilatenot expected the placeto delete of articulation in final ofposition. a following However, consonant. unlike We /t/, take the this coronal as evidence fricative

15and Final Hebrew [n] can alphabet also marginally letters; einstein surface["ajnstajn] beforeor the["ejnstejn] glide ,[j]. the energyThe only unit; Catalan and foreign words place with names a word-final like Main [-jn] ["mejn]cluster,or however, borrowings are like theairlines following["erlajns] three. A technical final cluster loanwords: like [-ln], whose ain [ˈ profileajn], or is also[ˈain], sonority-falling, zain [ˈzajn], isor not [ˈzain], found;Aramaic Lincoln and isHebrew pronounced alphabet as ["liNkon] letters;. However,einstein [ weˈajnstajn] do find or words [ˈejnstejn], with final the[-lm] energy, as in unit;elm ["Elm] and foreign‘type of place helmet’,names like(p)salm Main["salm] [ˈmejn]‘psalm’, or borrowingsfilm ["film] like‘film’. airlines [ˈerlajns]. A final cluster like [-ln], whose profile is also sonority-falling,16Recall that the velaris not nasal found; is always Lincoln the resultis pronounced of place assimilation as [ˈliŋkon]. to a velarHowever, stop and we is rendereddo findin words word-final with final [-lm], as in elm [ˈεlm] ‘type of helmet’, (p)salm [ˈsalm] ‘psalm’, film [ˈfilm] ‘film’. 16position due to the application of final cluster simplification. The reason why it is preserved is therefore distinct from Recall what that we the see with velar labial nasal and is palatal always nasals the and result has of to do place with assimilation sharing place to with a thevelar unassociated stop and velar is rendered stop. in word-final position due to the application of final cluster simplification. The reason why it is preserved is therefore distinct from what we see with labial and palatal nasals and has to do with sharing place with the unassociated velar stop. 17 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 17 of 23

[s] and the non-nasal coronal sonorant [l] never assimilate the place of articulation of a following consonant. We take this as evidence that /s/ and /l/ are specified for [coronal] place, as opposed to /n/ and /t/, and therefore they never require the licensing that /n/ requires.17 that /s/ and /l/ are specified for [coronal] place, as opposed to /n/ and /t/, and therefore they 17 3.3.5never Lexical require exceptions the licensing that /n/ requires. that /s/ and /l/ are specified for [coronal] place, as opposed to /n/ and /t/, and therefore they Somenever items require in the Catalan licensing surface that /n/ withrequires. a final17 [n] even though a stressed vowel precedes the3.3.5 nasal, Lexical as exceptions in [ˈsɔn] ‘sleep, sleepiness’ (4c). This word creates a minimal pair with [ˈsɔ]Some3.3.5 ‘sound’ Lexical items in(cf. exceptions Catalan [sunuɾiˈtat] surface ‘sonority’), with a final which[n] even undergoes though a regular stressed final vowel /n/ precedes deletion the after stressednasal, as vowels. in ["sOn] These‘sleep, are sleepiness’ clear cases (4c). of This lexical word createsexceptionality. a minimal Lexical pair with exceptions["sO] ‘sound’ in OT can(cf.Some be[sunuRi"tat] items easily in modeledCatalan‘sonority’), surface through which with undergoes lexically-indexed a final [n] regulareven though final constraints/n/ a stresseddeletion (Pater vowel after 2009). stressed precedes Lexically- vow- the nasal, as in ‘sleep, sleepiness’ (4c). This word creates a minimal pair with ‘sound’ indexedels. These constraints, are["sOn] clear casesor morpheme-specific of lexical exceptionality. constraints, Lexical exceptionsare general in constraints OT can["sO] be easilythat apply (cf. [sunuRi"tat] ‘sonority’), which undergoes regular final /n/ deletion after stressed vow- tomodeled a subset through of the lexically-indexed lexicon of a language. constraints In (Pater words 2009). such Lexically-indexed as [ˈsɔn], it is constraints, more important or els. These are clear cases of lexical exceptionality. Lexical exceptions in OT can be easily tomorpheme-specific avoid the floating constraints, nasal than are general to avoid constraints an unlicensed that apply /n/. to a subsetTherefore, of the lexiconwe propose of a that language.modeled through In words lexically-indexed such as ["sOn], it constraints is more important (Pater 2009). to avoid Lexically-indexed the floating nasal constraints, than to avoid or amorpheme-specific lexically-indexed constraints, constraint are *Floating general constraintsI outranks that apply *Unlicensed to a subset of[n], the whichlexicon of in a turn an unlicensed /n/. Therefore, we propose that a lexically-indexed constraint *FLOATINGI out- dominates unindexed *Floating by transitivity. The tableaux in (29) and (30) show rankslanguage. *UNLICENSED In words such[n], as which["sOn] in, it turn is more dominates important unindexed to avoid *F theLOATING floatingby nasal transitivity. than to avoid The how unindexed [ˈsɔ] and indexed [ˈsɔn] are selected as the most optimal candidates, tableauxan unlicensed in (29)/n/ and. Therefore, (30) show we how propose unindexed that a["sO] lexically-indexedand indexed ["sOn] constraintare selected *FLOATING as theI mostout- respectively,optimalranks *U candidates,NLICENSED despite respectively, [n]having, which the in despite same turn dominates havingunderlying the unindexed same representation underlying *FLOATING representation /sɔn/:by transitivity.the only/sOn/ difference: The the betweenonlytableaux difference the in (29) two between and is (30)the theindex. show two how is the unindexed index. ["sO] and indexed ["sOn] are selected as the most optimal candidates, respectively, despite having the same underlying representation /sOn/: the (29)only(29) differenceTableau between for for["sO] [ˈsɔ] the two is the index. /sOn/ *FLOATINGI *UNLICENSED[n] *FLOATING (29) Tableau for ["sO] F /sOn/ *FLOATINGI *UNLICENSED[n] *FLOATING C VF C V

a. Cs VO Cn V * F a. s O n * C VF C V b. Cs VO Cn V *! (30) Tableaub. s forO["sOn]n *!

/sOn/I *FLOATINGI *UNLICENSED[n] *FLOATING (30)(30) Tableau for for["sOn] [ˈsɔn] F /sOn/I *FLOATINGI *UNLICENSED[n] *FLOATING C VF C V a. Cs VO Cn V *! * F a. s O n *! * C VF C V

b. Cs VO Cn V * The finalb. OT grammars O n proposed in this paper is illustrated* in (31).

17 TheThis final interpretation OT grammar of the proposed facts, which in this is based paper on is the illustrated idea that some in (31). coronal consonants are specified for place ([s, l]) and others are not ([n, t]), does not necessarily conflict with OT’s Richness of the Base Hypothe- 17 This interpretation of the facts, which is based on the idea that some coronal consonants are specified sis.17 By assuming a set of feature co-occurrence markedness constraints of the type *[nasal,coronal] for [n] and for Thisplace interpretation ([s, l]) and ofothers the facts, are whichnot ([n, is basedt]), does on thenot idea necessarily that some conflict coronal consonantswith OT’s areRichness specified of forthe Base *[placeson, ( cont,coronal]) and others for are[t] not, an ( input), nasal does or not stop necessarily specified for conflict place with will never OT’s surface. Richness These of the features Base Hypothe- can only Hypothesis.− [s,− l] By assuming a set[n, of t]feature co-occurrence markedness constraints of the type *[nasal, coronal] co-occursis.for By[n] assumingand in place-sharing *[–son, a set –cont, of configurations feature coronal] co-occurrence for in order[t], an to markedness input avoid nasal unlicensed constraints or stop segments. specified of the This type for is *[nasal,coronal] achievedplace will if *UneverNLICENSED for surface.[n] and These dominates*[featuresson, cont,coronal]can *[nasal,coronal] only co-occur for [t] and, anin *[ inputplace-sharingson, nasalcont,coronal], or stop configurations specified which for in place turn in dominate willorder never to M surface.avoidAX[coronal]. unlicensed These features segments. can only This is − − − − co-occurachieved in if place-sharing *Unlicensed configurations dominates in *[nasal,coronal] order to avoid unlicensed and *[– segments.son, –cont, This coronal], is achieved which if *U inNLICENSED turn dominate dominatesMax[coronal]. *[nasal,coronal] and *[ son, cont,coronal], which in turn dominate MAX[coronal]. − − 18

18 Art. 63, page 18 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

The final OT grammar proposed in this paper is illustrated in (31).

(31)(31) Final OTOT grammargrammar as as a Hassea Hasse diagram diagram *NT#

DEP-V *UNENGAGED-CV *FLOATINGI FAITH-STRESS

*UNLICENSED[n]

*FLOATING

Having covered the native vocabulary of Catalan, we move on to discuss loanwords very briefly.

3.3.6Having The covered behavior the of native loanwords vocabulary of Catalan, we move on to discuss loanwords very briefly. As we noted in section 1, the process of final /n/ deletion extends to some loanwords (3), repeated3.3.6 The here behavior in (32). of loanwords18 As we noted in section 1, the process of final /n/ deletion extends to some loanwords (3), (32)repeatedFinal here in/n/ (32). deletion18 in loanwords a. Singular b. Plural c. Derivative (32) Final[uˈzo]/n/ deletion in loanwords[uˈzons] [uˈzɔnik] a. ‘ozone’Singular b. Plural c. Derivative‘ozonic’ [u"zo] [u"zons] [u"zOnik] [təliˈβa]‘ozone’ [təliˈβans] ‘ozonic’[təliβəˈnizmə] ‘Taliban’ ‘Talibanism’ [t@li"Ba] [t@li"Bans] [t@liB@"nizm@] [rəməˈða] [rəməˈðans] [rəməðəˈnεt] ‘Taliban’ ‘Talibanism’ ‘Ramadan’ ‘small Ramadan’ [siŋkɾuˈtɾo][r@m@"Da] [r@m@"Dans][siŋkɾuˈtɾons] [r@m@D@"nEt] ‘synchrotron’‘Ramadan’ ‘small Ramadan’ Nevertheless, as was illustrated in (4d), most loanwords do pattern with lexical exceptions in the sense that[siNkRu"tRo] final /n/ is realized.[siNkRu"tRons] In (33), we provide more such loanwords. We also add a few foreign‘synchrotron’ place names, as well as a few words that have entered the language more or less recently. Nevertheless, as was illustrated in (4d), most loanwords do pattern with lexical exceptions in the sense that final /n/ is realized. In (33), we provide more such loanwords. We also add a (33)few foreignFinal place /n/ names,deletion as blocked well as a in few loanwords words that have entered the language more or less recently.[məɣəˈzin] [uɾəŋguˈtan] [ʃərləsˈton] ‘magazine’ ‘orangutan’ ‘Charleston’ [ʃəˈman] [kəjˈman] [diˈβan] ‘shaman’ ‘caiman’ ‘divan, couch’ [tuβuˈɣan] [pəkisˈtan] [ˈklɔn] ‘slide’ ‘Pakistan’ ‘clone’ [duˈβlin] [berˈlin] [peˈkin] ‘Dublin’ ‘Berlin’ ‘Peking’ 18In Mascaro´ (1976), the word ozo´ ‘ozone’ is given as [u"zon], without final /n/ deletion, as opposed to the current standard[suˈðan] form. A reviewer[sinˈtɾon] notes that Mascar[fuˈton]o´ was probably using an older edition of Pompeu Fabra’s Diccionari general de la llengua catalana in which the spelling was ozon. This fact demonstrates the relatively unstable situation‘Sudan’ of loanwords‘Sintrom’ with respect to the process‘futon’ of final /n/ deletion over time.

18 In Mascaró (1976), the word ozó ‘ozone’ is given19 as [uˈzon], without final /n/ deletion, as opposed to the current standard form. A reviewer notes that Mascaró was probably using an older edition of Pompeu Fabra’s (1931) Diccionari general de la llengua catalana in which the spelling was ozon. This fact demonstrates the relatively unstable situation of loanwords with respect to the process of final /n/ deletion over time. Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 19 of 23

However, some of these words do undergo final /n/ deletion in colloquial Catalan, although not for all speakers, as already noted in Jiménez & Lloret (2011).19 We therefore contend that although final /n/ deletion is generally blocked in loanwords, the productivity of the process cannot be denied. This concludes our analysis of the exceptions to /n/ deletion. 4 Previous approaches Before concluding, we briefly review some previous work on final /n/ deletion in Catalan. Kikuchi (2002, 2009) proposed to relativize the margin sonority constraint hierarchy of Prince & Smolensky (1993/2004) according to the stressed syllable (34).

(34) Margin sonority constraint hierarchy relativized to stressed syllables *M-ˈσ/vowel >> *M-ˈσ/glide >> *M-ˈσ/liquid >> *M-ˈσ/nasal >> *M-ˈσ/obstruent

Ranking *M-ˈσ/nasal above the anti-deletion constraint Max-IO selects the candidate with final /n/ deletion only when the nasal is parsed as a margin of a stressed syllable. Otherwise, the candidate without final /n/ deletion is selected, becauseMax -IO dominates the more stringent constraint *M/nasal, which is not relativized to the stressed syllable. In order to block final /n/ deletion in plural forms, another constraint is introduced, Juncture-Contiguity, which requires adjacent segments across a morpheme boundary in the input to be adjacent in the output. This constraint outranks *M-ˈσ/nasal. Kikuchi’s (2002, 2009) approach suffers, however, from a fundamental problem. Stressed syllables are metrically prominent positions, and there is ample typological evidence that prominent positions are more faithful to the input than non-prominent positions, that is, prominent positions preserve contrast and avoid neutralization (Smith 2004, 2008). Relativizing the margin sonority constraint hierarchy for stressed syllables seems therefore unjustified from a typological perspective.20 Moreover, specifically in comparison to the present account, Kikuchi’s analysis references morphological boundaries, a move that our account manages to avoid. Bonet et al. (2004, 2005) suggested two possible ways to model what they see as a crazy rule: first, the use of parochial constraints like *ˈVn##, which prohibits a word- final configuration of a stressed vowel followed by [n]. However, the use of parochial constraints contradicts the universality of constraints postulated for OT, which should be simple and general. Second, they proposed to use the mechanism of local constraint conjunction and suggested the constraint Non-Finality&*n, which is violated by those outputs with final stress that contain [n]. Although this constraint receives a general formulation, it fails to explain why Non-Finality locally conjoins precisely with the general markedness constraint *n, and not, for instance, *k or *ʎ, which can be seen as more marked segments. The universality of this locally conjoined constraint is again dubious. Wheeler (2005) proposed instead an analysis based on allomorphy, where the allomorph without final /n/ is selected in final position, and the allomorph with final /n/ is selected elsewhere. However, such an analysis does not explain why there is no allomorphy and subsequent allomorph selection if stress is not final. In addition, the account is extremely

19 According to Zuraw (2016), in cases of exceptionality most words undergo a certain process, a smaller set of words shows the exceptional pattern, and an even smaller set of words behave variably. This situation is referred to as polarized variation, that is, a situation in which only a small number of items shows variation while the vast majority of items behaves in a consistent way. The situation found in Catalan for final /n/ deletion seems to be a case of this sort. It is not our purpose, however, to model this scenario: we refer the interested reader to Zuraw (2016). 20 On top of that, Kikuchi’s analysis cannot in principle block final /n/ deletion in [ˈkarn]-type cases, and does not account for the opaque [ˈden]-type cases, those counterfed by final cluster simplification. Art. 63, page 20 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

inelegant: for every alternating word, the analysis would have to assume two lexical representations which are identical in all respects except for the final /n/. Thus, several hundreds of cases with two stems differing only in the final consonant will have to be assumed for Catalan. In sum, it is clear that no previous account has provided a fully satisfactory explanation for the facts. By way of a conclusion, let us now enumerate the several advantages of our own account. 5 Conclusion In this paper we have carried out an analysis of the extent of final /n/ deletion in Catalan combining the representational assumptions and stress assignment principles of Strict CV and OT computation. The basic idea of our analysis is that final /n/ remains unassociated below its C position in order to satisfy a constraint against associated, but unlicensed nasals. The nasal can stay unassociated insofar as its CV is identified by metrical structure, as is the case when stress is final. When stress is non-final, however, the last CV unit of the phonological string is not engaged by either the melodic tier or the metrical tier. Therefore, in order to satisfy a constraint against unengaged CV structure, final /n/ is associated to its C position. Our analysis is simple and constitutes the first attempt to find a unified explanation for final /n/ deletion and its conditioning factor, stress. Under this analysis, phonological opacity disappears. Now we are in the position to give a concise answer to the research questions posed at the beginning of this paper:

(35) Research questions Q1. Why is /n/ the only nasal coda to be systematically deleted? And why only word-finally? A1. /n/ is the only coda to be systematically deleted because it is the weakest of consonants: it lacks a place node and therefore seeks to associate to the place node of a following consonant. In word-final position, however, the nasal is not followed by any other consonant to whose place node it could be associated. Q2. Why must stress immediately precede the deleted /n/? A2. When stress precedes the deleted /n/, the latter’s CV structure is still interpreted by the metrical structure. If the nasal does not immediately follow the stress, the CV unit is outside the domain of the foot. In these cases, the interpretability of the CV unit takes precedence over solving the problem posed by the unlicensed /n/, and the nasal ends up linked to its C position. Q3. Why is final /n/ deletion blocked when /n/ is the second element in a word-final consonant cluster? A3. For the same reason as Q2. In such a configuration, there is an empty nucleus standing between the two consonants. Because a foot’s head can only be placed on full vowels, the foot does not engage the last CV unit, and therefore the /n/ is linked to its C position in order to allow its interpretability. Q4. Why does final cluster simplification counterfeed final /n/ deletion? A4. This opaque interaction is solved if a containment theory of faithfulness is assumed. The unrealized obstruent is delinked from its C position but never deleted. The presence of a C-place node on that obstruent is sufficient to create a place-sharing configuration and therefore license the nasal. Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 21 of 23

Q5. What is the status of lexical exceptions? A5. Lexical exceptions in which final /n/ is not deleted even though it follows stress are dealt with by means of the lexically-indexed constraint

*FloatingI, which is ranked above the constraint against unlicensed nasals.

Beyond the specific issue of final /n/ deletion in Catalan, we emphasize that the success of our account lies in the link between skeletal position and stress assignment. It thus illustrates the claim that a skeleton-based metrical analysis is a worthy opponent to the mora-based one. From a more general perspective, we hope that this paper inspires other phonologists to pursue a compromise between the need to take serious care of representations, as allowed by GP or Strict CV, and the benefits of OT as a general theory of phonological computation through constraint interaction.

Abbreviations m = masculine, f = feminine

Acknowledgements We thank three anonymous reviewers for their helpful comments. Competing Interests The authors have no competing interests to declare. References Blevins, Juliette. 2004. Evolutionary phonology: The emergence of sound patterns. Cambridge: Cambridge University Press. DOI: https://doi.org/10.1017/CBO9780511486357 Blevins, Juliette. 2006. Syllable typology. In Keith Brown (ed.), Encyclopedia of language and linguistics, volume 12. 333–337. Oxford: Elsevier. DOI: https://doi.org/10.1016/ B0-08-044854-2/00047-X Bonet, Eulàlia & Francesc Torres-Tamarit. 2010. Allomorphy in pre-clitic imperatives in Formenteran Catalan: An output-based analysis. In Sonia Colina, Antxon Olarrea & Ana Maria Carvalho (eds.), Romance Linguistics 2009. Selected papers from the 39th Linguistic Symposium on Romance Languages, 337–352. Philadelphia/Amsterdam: John Benjamins. Bonet, Eulàlia & Maria-Rosa Lloret. 1998. Fonologia catalana. Barcelona: Ariel. Bonet, Eulàlia, Maria-Rosa Lloret & Joan Mascaró. 2004. Crazy rules and markedness in Optimality Theory. Paper presented at the Workshop on markedness in phonology, 27th Generative Linguistics in the Old World (GLOW27), Thessaloniki, 18–21 April 2004. Bonet, Eulàlia, Maria-Rosa Lloret & Joan Mascaró. 2005. How unnatural and exceptional can languages become? Paper presented at the phonology workshop Crazy rules and lexical exceptions, 3rd International Conference on Language Variation in Europe (ICLaVE3), Amsterdam, 23–25 June 2005. Cabré, Teresa. 1993. Estructura gramatical i lexicó: El mot mínim català. Catalonia: Universitat Aut’ónoma de Barcelona dissertation. Cabré, Teresa & Michael Kenstowicz. 1995. Prosodic trapping in Catalan. Linguistic Inquiry 26. 694–705. Clements, Nick & Elizabeth Hume. 1995. The internal organization of speech sounds. In John Goldsmith (ed.), Handbook of Phonological Theory, 245–306. Oxford: Basil Blackwell. Fabra, Pompeu. 1931. Diccionari general de la llengua catalana. Barcelona: Edhasa. Art. 63, page 22 of 23 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan

Gulsoy, Joseph. 1997. El tractament de la -n < N’ en català. Caplletra 20. 33–82. Harris, John & Edmund Gussmann. 1998. Final codas: Why the west was wrong. In Eugeniusz Cyran (ed.), Structure and interpretation—Studies in phonology, 139–162. Lublin: Folium. Harris, John & Edmund Gussmann. 2002. Word-final onsets. UCL Working Papers in Linguistics 14. 1–42. Jiménez, Jesús & Maria-Rosa Lloret. 2011. Hòmens, homes i conflictes de naturalitat. In Josep Massot i Muntaner (ed.), Estudis de llengua i literatura catalanes/LXIV. Miscel.lània Albert Hauf, 1, 217–245. Barcelona: Publicacions de l’Abadia de Montserrat. Kaye, Jonathan. 1990. ‘Coda’ licensing. Phonology Yearbook 7. 301–330. DOI: https://doi. org/10.1017/S0952675700001214 Kaye, Jonathan, Jean Lowenstamm & Jean-Roger Vergnaud. 1990. Constituent structure and government in phonology. Phonology Yearbook 7. 193–231. DOI: https://doi. org/10.1017/S0952675700001184 Kikuchi, Seiichiro. 2002. Positional markedness in Catalan word-final deletion. Paper presented at the 125th meeting of the Linguistic Society of Japan, Tohoku Gakuin University, 4 November, 2002. Kikuchi, Seiichiro. 2009. Stress-conditioned morphophonological asymmetries in Spanish, Catalan and Galician. Paper presented at the Phonology Forum 2009, Kobe University, 24–25 August, 2009. Kiparsky, Paul. 1985. Some consequences of Lexical Phonology. Phonology Yearbook 2. 85–138. DOI: https://doi.org/10.1017/S0952675700000397 Levelt, Claartje & Marc Van Oostendorp. 2007. Feature co-occurrence constraints in L1 acquisition. Linguistics in the Netherlands 24. 162–172. DOI: https://doi.org/10.1075/ avt.24.16lev Lowenstamm, Jean. 1996. CV as the only syllable type. In Jacques Durand & Bernard Laks (eds.), Current trends in phonology models and methods, 419–442. Salford: European Studies Research Institute, University of Salford. Mascaró, Joan. 1976. Catalan phonology and the phonological cycle. Cambridge, MA: Massachusetts Institute of Technology dissertation. Pater, Joe. 2009. Morpheme-specific phonology: Constraint indexation and inconsistency resolution. In Steve Parker (ed.), Phonological argumentation: Essays on evidence and motivation, 123–154. London: Equinox. Polgárdi, Krisztina. 1998. Vowel harmony: An account in terms of Government and Optimality. Leiden: Universiteit Leiden dissertation. Prince, Alan & Paul Smolensky. 1993. Optimality Theory: Constraint interaction in generative grammar. Unpublished manuscript, Rutgers University & University of Colorado, Boulder. Published 2004, Malden, MA & Oxford: Blackwell. Rowicka, Grace. 1999. On ghost vowels: A strict CV approach. Leiden: Universiteit Leiden dissertation. Sampson, Rodney. 1999. Nasal vowel evolution in Romance. Oxford: Oxford University Press. Scheer, Tobias. 2004. A lateral theory of phonology: What is CVCV, and why should it be? Berlin: Mouton de Gruyter. DOI: https://doi.org/10.1515/9783110908336 Scheer, Tobias & Péter Szigetvári. 2005. Unified representations for stress and the syllable. Phonology 22. 37–75. DOI: https://doi.org/10.1017/S0952675705000436 Serra, Josep. 1996. La fonologia prosòdica del català. Girona: Universitat de Girona dissertation. Smith, Jennifer. 2004. Making constraints positional: Toward a compositional model of Con. Lingua 114. 1433–1464. DOI: https://doi.org/10.1016/j.lingua.2003.08.003 Faust and Torres-Tamarit: Stress and final /n/ deletion in Catalan Art. 63, page 23 of 23

Smith, Jennifer. 2008. Markedness, faithfulness, positions, and contexts: Lenition and fortition in Optimality Theory. In Joaquim Brandão de Carvalho, Tobias Scheer & Philippe Ségéral (eds.), Lenition and fortition, 519–560. Berlin: Mouton de Gruyter. Torres-Tamarit, Francesc, Kathrin Linke & Maria del Mar Vanrell. 2017. Opacity in Campidanian Sardinian metaphony. Natural Language and Linguistic Theory 35. 549–576. DOI: https://doi.org/10.1007/s11049-016-9341-0 Ulfsbjorninn, Shanti. 2014. A field theory of stress: The role of empty nuclei in stress systems. London: University of London dissertation. Vallverdú, Teresa. 1997. On stress assignment in Catalan. Barcelona: Universitat Autònoma de Barcelona dissertation. VanDam, Mark. 2004. Word final coda typology. Journal of Universal Language 5. 119–148. DOI: https://doi.org/10.22425/jul.2004.5.1.119 Van Oostendorp, Marc. 2014. Representing variation: The view from phonological theory. Paper presented at the Theoretical issues in contemporary phonology: Reading Tobias Scheer, February 6–8, Paris. Wheeler, Max. 2005. The phonology of Catalan. Oxford: Oxford University Press. Zuraw, Kie. 2016. Polarized variation. In Eulàlia Bonet & Francesc Torres-Tamarit (eds.), Catalan Journal of Linguistics 15. 145–171.

How to cite this article: Faust, Noam and Francesc Torres-Tamarit. 2017. Stress and final /n/ deletion in Catalan: Combining Strict CV and OT. Glossa: a journal of general linguistics 2(1): 63. 1–23, DOI: https://doi.org/10.5334/gjgl.64

Published: 30 June 2017

Copyright: © 2017 The Author(s). This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. See http://creativecommons.org/licenses/by/4.0/.

Glossa: a journal of general linguistics is a peer-reviewed open access journal

published by Ubiquity Press. OPEN ACCESS