DOCSLIB.ORG

From Latin to Romance

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
From Latin to Romance From Latin to Romance: Computational Modeling of Syncretism Tyler Lau∗, Maria Polinsky†, and Jake Seaton† ∗Department of Linguistics, University of California at Berkeley, USA †Deparment of Linguistics, Harvard University, USA Overview Structure of the Connectionist Model Discussion • What factors in Late Latin led to the heavy reshaping of • With phonology, frequency,& human semantics /kanis/ ‘dog’ Corpus the nominal system? 454 Latin Vulgate nouns in 6 forms (3 cases 2 numbers) = 2724 total forms • Declensions IV & V fall out in every simulation Phonology Semantics # times each form introduced = log(freq(form))× relative freq(case & num combination) • What minimal information does a connectionist model In training, input of each token (phonology and semantics)× and expected output given to model • With case hierarchy added, final forms converge more a ... and hidden layer adjusts connection weights accordingly need to predict syncretism in the correct direction? k Phonology (396 nodes = 6 6 11) Human Semantics (8 nodes) • Genitive singular drops out completely [-human] × × Analogy driven by factors such as frequency, [-vd] [+C] ... [+vd] [-C] ... ... Each word maximally 6 syllables Nouns coded as: • Genitive plural hardly survives (only example in • Each syllable maximally 6 phonemes (CCVVCC) Male human: first 4 nodes activated Each phoneme coded for 11 features markedness, and morpheme length. (Kuryłowicz 1947, -1 1 ... 1 -1 ... ... 0 0 0 0 0 0 0 0 Female human: final 4 nodes activated history is oblique 3PL pronoun–Fr. leur, It. loro) Bybee 1985, Albright 2008) Non-human: 0 nodes activated • Forms remaining in ≥90% of simulations Hidden Layer (30 nodes): -am > -a F.SG ending in all Romance (> -e in Fr.) Learning takes place in the hidden layer (Goldsmith & O’Brien 1995) • • Changes in Romance have been attributed not only to ... ? ? ? ? ? Hidden nodes use info from input and output to adjust connection weights between input layer • -um > -u M.SG ending in all of Romance (> -o in Sp., It. etc.) and hidden layer and between hidden layer and output layer sound change, but also to contact • -em > -e SG ending for M/F nouns in all of Romance Output Layer (13 nodes): Two parameters to be toggled: We aim to use a connectionist simulation of generational 1) CASE HIERARCHY • Forms remaining in 25-90% of simulations • 1 0 0 0 0 1 0 0 1 0 0 1 0 2) GENITIVE DROP learning providing minimal phonological and semantic No: All cases coded as equidistant No: All three cases available as output • -? SG ending for M/F nouns in all of Romance nom = (1,0,0) acc = (0,1,0) gen = (0,0,1) Possible outputs: nom, acc, gen nom sg • -¯es PL ending in western Romance, maybe > -i in eastern information and see whether the changes that are M III Yes: acc equidistant between nom and gen Yes: gen output becomes unavailable nom = (1,0,0) acc = (1,1,0) gen = (1,1,1) Possible outputs: nom, acc • -¯os M.PL ending in western Romance, maybe > -i in eastern actually attested in Romance can be reproduced Gender Declension Case Number Generational Learning • -¯as F.PL ending in western Romance, maybe > -e in eastern Output set of features becomes new expected output for following generation • M/F.NOM.SG -us & -as: in E-Romance., final -s falls out; in Next Generation Iterated for 15 generations Latin Romance Comments W-Romance, nom persists in older Sp. & Fr. I Results • Accusative is most robust form–in history, acts as base II I form in most of modern Romance. Plurals in some Declensions IV and V languages may be from nominative (D’hulst 2006) Declension III II merge to III and I early • With genitive dropped, two notable outcomes for neuter IV III NSG to MSG NPL to FSG NPL to FPL 1.00 1.00 1.00 1 N.SG > M.SG | N.PL > F.SG (most of Romance) V 2 M.SG > N.SG | N.PL > F.PL (Romanian system) In most of Romance, ev- erything merges to the 0.75 0.75 0.75 nom accusative. The vocative • Taking into account these minimal factors, simulation falls out of use as well. offers a rather accurate history of syncretism and trends Nominative survives in acc nom early Spanish and French. 0.50 0.50 0.50 Also possible that East- that occurred on way to modern Romance languages ern Romance plurals gen acc come from nominative plurals 0.25 0.25 0.25 % of NPL to FPL Genitive and dative sur- % of NPL to FSG Case % of NSG to MSG References dat gen vive only for feminine nouns in Romanian, vocative survives only 0.00 0.00 0.00 abl voc for masculine 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Albright, A. 2008. Explaining Universal Tendencies and Language Particulars in Analogical Change. In Generation Generation Generation Linguistics Universals and Language Change, 144-181. Bybee, J. 1985. Morphology: A Study of the voc Relation between Meaning and Form. D’hulst, Y. 2006. Romance Plurals. Lingua 116, 1303-1329. In most of Romance, With genitive drop, neuter singu- With genitive drop, neuter plurals Comparing with Figure 3, it is no- Goldsmith, J. & O’Brien, J. 1995. Grammar within a Neu- neuters mostly enter the Figure 2: Figure 3: Figure 4: ral Net. In The Reality of Linguistic Rules, 95-113. m masculine singular class lars bifurcate–they either merge with mascu- almost consistently migrate to the feminine table that in those trials where neuter plu- due to similar phonology. Herman, J. 1967. Le Latin Vulgaire. m f Due to similar phonology lines or draw masculines to their class (see singular class due to phonological similarity rals do not migrate to the feminine singular Kuryłowicz, J. 1947. The Nature of the so-called Analogical Processes. Diachronica 12 (1), with the feminine plural Figure 6). alone. class, they migrate to the feminine plural. 113-145. and fact that neuter plu- Gender f m.sg rals can be interpreted as collectives, some neuters Forms Remaining at End of Simulation (w/ Hierarchy) Distribution of Genders by End of Simulation Distribution of Cases by End of Simulation Acknowledgements become feminine 1.0 n.sg Gender: n m f Case: acc nom gen Romanian has an ambi- 0.9 f.sg generic neuter gender– 1.0 1.0 neuter singulars take 0.8 0.9 0.9 Many thanks to Kevin Ryan, James Kirby, Andrew Garrett, Terry Regier, n.pl masculine morphology 0.7 f.pl while neuter plurals take 0.8 0.8 Mairi McLaughlin, and Yang Xu for comments and guidance, to Ezra Van feminine morphology 0.6 0.5 0.7 0.7 Everbroeck for providing the code for the simulation in Polinsky and Van 0.4 0.6 0.6 Everbroeck (2003), and to Edwin Ko for consultation on data visualization. 0.3 0.5 0.5 0.2 0.4 0.4 Latin Declension System Proportion of Trials Percentage 0.1 0.3 Percentage 0.3 Contact Information 0.0 0.2 0.2 0.1 0.1 • Tyler Lau: [email protected] I II IIIa IIIb IV V f I acc pl f I gen pl f I acc sg f I nom sg f III accf I nom pl pl f III genn II accfpl II acc pl sgm I acc pl n II accm sg II accm sg III acc pl m II facc III acc pl f sgIII nom sg m IIm gen IIIm genpl II nnom plII nomn plIII accsg sg f II nom nsg II nom pl 0.0 0.0 m III accm sg II nom sgm III nom sg m III nom pl n III nom sg Root silva- anno- color- igni- lacu- fide- Form Trial Trial • Maria Polinsky: [email protected] Gloss ‘forest’ ‘year’ ‘color’ ‘fire’ ‘lake’ ‘faith’ • Jake Seaton: [email protected] Nom. silva annus color ignis lacus fid¯es Figure 5: With case hierarchy in play, ac- Figure 6: In most of the trials, the neuter Figure 7: With case hierarchy taken into Sg. Gen. silvae ann¯ı col¯oris ignis lacus¯ fide¯ı cusative is very robust and only the genitive falls out. In the cases where it is more consideration, the accusative becomes the Acc. silvam annum col¯orem ignem lacum fidem plural survives in some trials. Without it, the robust, masculine nouns migrate to the dominant case in most trials and the only Nom. silvae ann¯ı col¯or¯es ign¯es lacus¯ fid¯es genitive (both singular and plural) survive to neuter class and the total proportion of mas- case in almost half of them. The genitive Pl. Gen. silv¯arum annorum¯ col¯orum ignium lacum¯ fid¯erum a greater extent. culine+neuter nouns remains approximately survives in hardly more than 10% of trials Acc. silv¯as annos¯ col¯or¯es ign¯ıs/ lacus¯ fid¯es equal. (without case hierarchy, the genitive remains ign¯es in over 50% of trials). Figure 1: The Latin Declension Classes.
Load more

Recommended publications
  • PART I: HOMEWORK 1: GRAMMAR from the Book. 2: EXERCISES
    1 Winter Semester 2015/2016 Handout 2 PART I: HOMEWORK 1: GRAMMAR from the book. To be prepared for the Lesson 2 you are expected: to read the following material from the book Unit 1: Pronunciation and accent in Latin, p. 1-2; Grammatical categories, p. 3-5; 2: EXERCISES 1. Read aloud: • vertebra, ante, palpebra, medulla, vēna, trachēa, venēnum (2) • sine, pilula, vitrum, inter, spīna, rīma, vīnum, salīva (3) • post, anodus, oleum, prostata, bōlus, prō, prōcessus, dolorōsus (4) • apud, gutta, glandula, uterus, ūrīna, rūptūra, nātūra (5) • aegrōtus, praemātūrus, lagoena, foetor, aër, dyspnoē, diploē, proerythroblastos, coenzymum (6) • felleus, balneum, āreola, aorta, interosseae, pleura, pӯogenēs, euryōpia (9) • celulla, cibus, caecum, cystis, costa, cutis, fasciculus, clāvicula, frāctūra (11) • coccӯgeus, occipitālis, ōscilococcinum, accessōrius, saccus, saccī, vaccīna (12) • caecum, caecī, bucca, buccae, verrūca, verrūcae, thōrācica, thōrācicae, saccus, saccī, coenzymum (13) • digitus, tībia, destillāta, hernia, tunica, audītus (15) • fūnctiō, articulātiō, vitium, īnsufficientia, sānātiō, ōstium, testium, mixtiō, combustiō (16) • aqua, liquor, quadrātus, lingua, sanguis, unguentum, unguis, unguium, inguinālis (17) • resistentia, incīsūra, spongiōsus, basis, crisis, nasālis, pulsus, morsus, mēnsis, plasma (18) • comissūra, prōcessus, scissus, accessōrius, ossa, ossium, hypoglōssus, tussis, pertussis (19) 2. Read aloud the nominative and genitive forms of the nouns. Write down the number of the declension; follow the example: ex: caput,
    [Show full text]
  • Old French and Romanian Declensions from a Word and Paradigm Perspective and the Notion of “Default Syncretism”
    OLD FRENCH AND ROMANIAN DECLENSIONS FROM A WORD AND PARADIGM PERSPECTIVE AND THE NOTION OF “DEFAULT SYNCRETISM” ALAIN KIHM1 Abstract. Old French and Romanian nominal inflections or declensions share the property of being apparently uncomplicated as their paradigms consist in only two forms at most: a base form and an inflected form. This outward simplicity, however, results from complex syncretisms. In Old French masculine nouns, the singular subject case and the plural object case are identically inflected, whereas the singular object and the plural subject cases are identical base forms; in Romanian feminine nouns, the singular genitive-dative and the two plural case forms are the same. Such syncretisms raise a descriptive and theoretical issue as they appear to be neither semantically motivated nor fully arbitrary. Drawing on the conceptual and formalizing resources of Word and Paradigm (WP) theory and Paradigm Function Morphology (PFM), the present essay attempts to solve the issue by assuming a third kind of syncretism that involves not the meaningful content of features, but their DEFAULT value. At the same time, it proposes a nearly full treatment of Old French and Romanian declensions in PFM terms. Keywords: case, default, declension, gender, number, paradigm, syncretism. 1. INTRODUCTION As is well-known, only Old French and Romanian among Romance languages kept something of the Latin rich nominal inflection for case and number or DECLENSION. The present study aims to provide a synchronic account of this phenomenon in a Word and Paradigm (WP) framework (Blevins 2006), using the formal apparatus of Paradigm Function Morphology (PFM – Stump 2001; Bonami, Stump to appear).
    [Show full text]
  • Illinois Classical Studies
    14 Is it Really the Accusative? A Century-Old Controversy Revisited PAUL A. GAENG The stages involved in the disintegration of the classical Latin system of declensions and its evolution during the centuries that preceded the "birth" of the Romance languages have been adequately outlined by leading Romanists of both past and present. The undisputed master of Romance linguistics in our century, Wilhelm Meyer-Liibke, summed up the opinion of his generation when he insisted on the Latin accusa- tive as being the Romance "Normalkasus," with due allowances for the Latin nominative as reflected, for instance, in the cas sujet of Old French and Old Provenfal and the plurals of Italian and Rumanian nounsJ Anchored in the Diezian theory of the Latin accusative as the progenitor of the Old French and Old Proven9al oblique case and the single case forms of the other Romance languages, Meyer-Liibke's view that, except for sporadic instances of nominative derivation, the Romance noun is, in essence, a survival of classical Latin accusative forms both in the singular and the plural has generally prevailed, despite an occasional voice offering convincing arguments to the con- trary. The first scholars on record to challenge this "accusative theory" were the Italians D'Ovidio and Ascoli. The former, the catalyst for the subsequent declensional combat waged by Ascoli and Meyer-Liibke, set out to show in his SuU'origine delVunica flessione del nome (1872), that the post-classical form servo comprised not a single case but the classi- cal nominative servus, dative/ablative servo, and accusative servum, in the singular, and that the plural servi represented classical nominative servi and the dative/ablative servis.
    [Show full text]
  • On Latin Nominal Inflection: the Form-Function Relationship
    Published in: Morphology vol. 23 (2013) no. 2, p. 179-200. On Latin nominal inflection: the form-function relationship Bernd Wiese Abstract The present paper provides a new approach to the form-function relation in Latin declension. First, inflections are discussed from a functional point of view with special consideration to questions of syncretism. A case hierarchy is justified for Latin that conforms to general observations on case systems. The analysis leads to a markedness scale that provides a ranking of case-number-combinations from un- marked to most marked. Systematic syncretism always applies to contiguous sections of the case-number-scale (‘syncretism fields’). Second, inflections are analysed from a formal point of view taking into account partial identities and differences among noun endings. Theme vowels being factored out, endings are classified on the basis of their make-up, e.g., as sigmatic endings; as containing desinential (non-thematic) vowels; as containing long vowels; and so on. The analysis leads to a view of endings as involving more basic elements or ‘markers’. Endings of the various declensions instantiate a small number of types, and these can be put into a ranked order (a for- mal scale) that applies transparadigmatically. Third, the relationship between the in- dependently substantiated functional and formal hierarchies is examined. In any de- clension, the form-function-relationship is established by aligning the relevant formal and functional scales (or ‘sequences’). Some types of endings are in one-to-one cor- respondence with bundles of morphosyntactic properties as they should be according to a classical morphemic approach, but others are not.
    [Show full text]
  • The Latvian Declension*
    The Latvian declension* MORRIS HALLE 1. INTRODUCTION In this paper I examine in some detail the declension of Latvian nouns and adjectives in the light of a descriptive framework for morphology that I have developed in a number of recent papers (Halle 1990a, 1990b). The data examined provide clear evidence for the need to distinguish gender from inflection class, a distinction that is also supported by the facts from other languages reviewed in this volume in the papers by Harris, Hoberman, and Aronoff. In addition - and perhaps more importantly - the Latvian facts illuminate the central role played in the morphology by - what I have called - abstract morphemes and shed interesting new light on the ordering of rules, in particular, on the issue ofdisjunctive rule order and default rules. The flow chart in (0) represents the major components of the framework and their interrelations. (0) ,-----------, .--- 1 I : Morphology IMorphemes I ~ : ~ _ /' /' /' ". I ~justmenl I Words 1~-----0-----B ~;=::-l L-.- \lJ\ "" t + 1~1I~1 \ "t I Vocabulary \ ....~ I \ LJ I \ I \ Syntax - Semantics Phonology '- J The morphology portion of the framework represented by the two blocks on the right of (0) is in the main a formalization of the traditional view that words are composed of one or more morphemes and that pluri-morphemic words have the familiar nested constituent structure. The framework treats each morpheme as a complex symbol consisting of an identifying index paired with a set of markers representing the grammatical and semantic properties of the word. Although at first sight this may appear as an innova­ tion, it is but a modernized version of Saussure's widely discussed conception of the word as a sign composed of a signifier and a signified, which itself has ancient roots in Stoic philosophy (see Jakobson 1966).1 The identifying index of most morphemes consists of a sequence of 33 G.
    [Show full text]
  • Basic Medical Terminology
    Basic medical terminology 2 ND SEMINAR Read aloud Infarctus myocardii recens Fractura comminutiva colli femoris lateris dextri Commotio cerebri Vulnus punctum thoracis ad pulmonem lateris sinistri penetrans Aethylismus chronicus Appendicitis acuta Intoxicatio carboneo hydroxydato (CO) gradus maioris Read aloud Typhus reccurens Tonsillitis purulenta recidivans Diabetes mellitus stabilis Dermatitis allergica protrahens Vitium cordis acquisitum Infarctus haemispherii sinistri cerebri, Hemiparesis Nephrolithiasis, colica renalis subsequens Tumor ventriculi ad investigationem et observationem Questions How do I decide to what declension the word belongs? I need to know the genitive ending What are the genitive endings of Latin declensions? Questions What is a stem? a stem is a form to which affixes can be attached How can we identify the stem of a Latin noun? we take genitive form and we remove the genitive ending vena, ven-ae musculus, muscul-i; diameter, diametr-i auris, aur-is; corpus, corpor-is arcus, arc-us; genu, gen-us facies, faci-ei Gender The grammatical gender of a noun affects the form of other words related to it. In Latin, adjectives (and pronouns) change their form depending on the noun to which they refer. ENGLISH has 3 genders: HE – refers to male humans and animals SHE – refers to female humans and animals IT – inanimate objects or animals LATIN has 3 genders: not only humans and animals, but also other objects are thought of as being: masculine -> discipulus (he-student), musculus (muscle) feminine -> discipula (she-student), vena (vein) neuter -> corpus (body) Latin – inflectional language In many languages, Latin and Greek among them, nouns inflect (change their form) for number and for case.
    [Show full text]
  • Short Historical Latin Grammar
    t44:9^ SHORT HISTORICAL LATIN GRAMMAR r/' W. M. LINDSAY, M.A. t » » FELLOW OF JESUS COLLEGE, OXFORD AT THE CLARENDON PRESS 1895 bonbon HENRY FROWDE Oxford University Press Warehouse Amen Corner, E.C. Qten? ^orft MACMILLA.N & CO., 66 FIFTH AVENUE PREFACE Teachers of Latin Grammar have for a long time felt the need of a book which will exhibit the his- torical development of Latin Accidence and explain the anomalies of Latin Declension and Conjugation^ which will explain, for example, how itineris became the Genitiv(J of iter, how volo, vis, vult differ from lego, legis, legit^ why the Comparative of magnijicus should be the circum should have magnificentior, why Preposition a by-form circa. In this 8hort Historical Latin Grammar, designed for the Universities and the Higher Forms of Schools, I have tried to present this information in an intelligible and, if possible, interesting form. While making full use of the discoveries of Comparative Philo- logy, which have in recent years added so much to our knowledge of Latin, I have avoided the technical vocabulary of that science, and in quoting parallels to Latin words have restricted myself to the Greek, to the exclusion of Sanscrit, Gothic, and the other Indo- European languages. It is true that each and every problem of the Latin language has not yet been solved, but for all that the stability of most of the results reached by the methods of Comparative Philology is beyond question; and every one who has studied the ^3 vi Preface. subject witli any minuteness knows which results are certain and which may have to be modified by subsequent research.
    [Show full text]
  • German Kraus
    Kraus 1 Kelsey Kraus Ling 105 Morphology 19 March 2012 An in the German Language Type of Problem German is a language spoken natively in Western Europe by approximately 100 million people. It is a Western Germanic language, and is closely related to Dutch, Afrikaans, English, and the Scandinavian languages. Any student of German must learn a few crucial facts before beginning an analysis of the language. First off, let’s clarify some questions regarding German orthography and phonology. The German language dictates that nouns be capitalized in all positions throughout the sentence. In contrast to English, this means that proper nouns and regular nouns are not distinguished simply by capitalization. This will not be extremely important here, but this knowledge will facilitate an ease of interpretation of the data to come. The German consonants are pronounced the same as their English counterparts with two exceptions. The orthographic consonant cluster /ch/ has two realizations in German, [x] and [ç]. Forms pronounced like Fach ‘compartment’ take the [x] phoneme /fa:x/ and those like ich ‘I,’ are pronounced with the cedilla, /ɪ:ç/. There is a generalization that the [ç] phoneme appears after front vowels, and that [x] occurs after back vowels. This will not be a major part of this problem, but the phonology may factor in at some point. German also has a few umlauted vowels, whose phonetic transcriptions I have included below, and some diphthongs. Unless otherwise stated, German vowel orthography generally corresponds to the IPA symbol. ä ü ö ei au eu ɛ: y: ø: aɪ aʊ ɔʏ German also has three genders for its nouns: Masculine, Feminine and Neuter.
    [Show full text]
  • Both Latin and Polish Are Inflected Languages and the Use of Nouns in a Sentence Is Indicated Not by Position (As It Is in English) but by Case Ending
    Both Latin and Polish are inflected languages and the use of nouns in a sentence is indicated not by position (as it is in English) but by case ending. Word position in a Latin or Polish sentence does not necessarily indicate how the word is being used in the sentence. It is the ending that is of paramount importance. The following Latin sentences all have the same meaning regardless of word order. 1) Puer puellam pulchram vidit. 2) Puer vidit puellam pulchram. 3) Pulchram puellam puer vidit. 4) Vidit pulchram puellam puer. All mean: The boy saw the pretty girl. Polish Cases Latin Cases English Use Nominative Nominative Subject Genitive Genitive Possession Dative Dative Indirect Object Accusative Accusative Direct Object, Object of Prepositions, etc. Vocative Vocative Direct Address Locative* Ablative* Multiple uses Instrumental* Multiple uses *The ablative in Latin combines the uses of the earlier Latin Locative & Instrumental Cases and functions like the Polish Locative & Instrumental Cases. LATIN DECLENSION CASE ENDINGS Case 1st Decl. 2nd Decl. 3rd Decl. 4th Decl. 5th Decl. SINGULAR Nom. a us, er,ir,um var. us, u es Gen. ae i is us ei Dat. ae o i ui, u ei Acc. am um em, im um, u em Voc. Generally the same as the nominative except 2nd Decl. Nouns in us—Voc. Ends in e Abl. a o e, i u e PLURAL Nom. ae i, a es, a us, a es Gen. arum orum um uum erum Dat. is is ibus ibus ebus Acc. as os, a es, a us, a es Voc.
    [Show full text]
  • Dissertation Master Document-1
    Copyright by Sarah Elizabeth Harmon 2007 The Dissertation Committee for Sarah Elizabeth Harmon Certifies that this is the approved version of the following dissertation: Gender in the Romance Languages: An Evolutionary Approach Committee: Brigitte L.M. Bauer, Supervisor Carlos A. Solé Frederick Hensey Cinzia Russi Qing Zhang Gender in the Romance Languages: An Evolutionary Approach by Sarah Elizabeth Harmon, BA; MA Dissertation Presented to the Faculty of the Graduate School of The University of Texas at Austin in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy The University of Texas at Austin August 2007 Dedication To my parents, Wilmoth C. (Jr.) and Michelle J. Harmon, for their tireless enthusiasm and unwavering love, and my brothers Matthew and Mark for their constant support. In memory of Professors Stephen LaPointe and Mark R.V. Southern Acknowledgements Language has always fascinated me. Growing up in a partially- multilingual family and in a highly diverse and multilingual area, I have always tried to imitate the sounds, words and phrases around me. Once I started studying Spanish, I was comparing it to English. Once I started learning Italian, I began to compare it to dialectal Italian phrases spoken in my family, and to Spanish and English. Little did I know that I had been essentially performing contrastive analyses that whole time, albeit in a crude fashion. It was not until I took a required ‘Introduction to Linguistics’ course that I truly found my calling. The questions posed—and answered—in this dissertation reflect many years of casual and formal research: why do certain aspects of languages change, but not others? Why are there differences among some dialects in a given language? Why are certain changes so similar to one another? I think it is fair to say that we all know that languages change, but few can say why they do so.
    [Show full text]
  • The Revised Latin Primer
    REVISED [ ATTM O T> T \ f T?"D JUAI UN i RIMcK UCSB LIBRARY 74. THE REVISED LATIN PRIMER BT BENJAMIN HALL KENNEDY, D.D. FORMERLY FKLLOW OS ST. JOHN'S COLLEGE, CAMBRIDGE REGIUS PROFESSOR OF GREEK A.ND CANON Or ELY TWELFTH IMPRESSION LONGMANS, GKEEN, AND CO. 39 PATERNOSTEK ROW, LONDON NEW YORK AND BOMBAY 1906 All rights reserved By the same Author. THE SHORTER LATIN PRIMER. Fcp. 8vo. price ONE SHILLING. AUTHOR'S PREFACE. Tnis little boofc is published in compliance with a wish expressed to me by a large number of Masters, including many teachers in Preparatory Schools, and in the lower forms of Public Schools. It is intended to be a simple manual for beginners in Latin, preparatory to the use of the Revised Latin Primer. In order to facilitate the passage of the learner from one book to the other, and also to make it possible for them to be used side by side if necessary, this Shorter Primer has been made, as far as it goes, in the main identical with the Revised Primer, and arranged on the same plan. It contains the memorial portion of the Accidence, with a few of the most important notes and explanations, and the more elementary parts of the Syntax of the Simple Sentence. A short outline of the Compound Sentence, which did not form part of my original plan, has been added in accordance with the desire of several experienced teachers, In this part it has been necessary, for the sake of brevity, to make the arrangement and wording somewhat different from that of the Revised Primer, In this book, as in the Revised Primer, I have endeavoured to simplify the marking of quantity by placing the mark of quantity, as a rule, only on the long vowels.
    [Show full text]
  • Latin for Beginners
    Latin for Beginners Author: Benjamin Leonard D'Ooge Project Gutenberg's Latin for Beginners, by Benjamin Leonard D'Ooge This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at www.gutenberg.org Title: Latin for Beginners Author: Benjamin Leonard D'Ooge Release Date: April 25, 2006 [EBook #18251] Language: English Character set encoding: UTF-8 *** START OF THIS PROJECT GUTENBERG EBOOK LATIN FOR BEGINNERS *** Produced by Louise Hope, Dave Maddock and the Online Distributed Proofreading Team at http://www.pgdp.net (This file was made using scans of public domain material by TextKit.com) *Transcriber’s Notes: This textbook uses many letters with macrons (straight line above vowel): ā ē ī ō ū ȳ Ā Ē Ī Ō Ū If these characters do not display properly--in particular, if the macrons do not appear directly above the letters--you may have an incompatible text reader or unavailable fonts. See if you can change the “character set” or “file encoding” to Unicode (UTF-8), or try a different font. As a last resort, you may need to use one of the simplified versions of this file: Latin-1 (Windows), Mac, or ASCII-7 (“typewriter” characters only). Other combined forms occur much less often: ă ĕ ĭ ŏ ŭ (single letter with breve) oo (“oo” with breve) ē ī( letter with co bined breve and acron) ā ī ĕ ĭ ŏ ŭ (letters with co bined breve acron and accent used only in Pronunciation section) Note that most accents will correctly appear _after_ a letter: īn´sula ha´bitat illī´us To make this unpaginated e-text easier to use each chapter’s Special Vocabulary has been included with its chapter _in addition to_ its original location near the end of the book.
    [Show full text]