OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi Understanding and Measuring Morphological Complexity OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi Understanding and Measuring Morphological Complexity Edited by MATTHEW BAERMAN, DUNSTAN BROWN, AND GREVILLE G. CORBETT 1 OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi 3 Great Clarendon Street, Oxford, ox2 6dp, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © editorial matter and organization Matthew Baerman, Dunstan Brown, and Greville G. Corbett 2015 © the chapters their several authors 2015 Themoralrightsoftheauthorshavebeenasserted First Edition published in 2015 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2014947260 ISBN 978–0–19–872376–9 Printed and bound by CPI Group (UK) Ltd, Croydon, cr0 4yy Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work. OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi Contents List of Figures and Tables vii List of Abbreviations xi List of Contributors xiii Part I. What is Morphological Complexity? Understanding and measuring morphological complexity: An introduction Matthew Baerman, Dunstan Brown, and Greville G. Corbett Dimensions of morphological complexity Stephen R. Anderson Part II. Understanding Complexity Rhizomorphomes, meromorphomes, and metamorphomes Erich R. Round Morphological opacity: Rules of referral in Kanum verbs Mark Donohue Morphological complexity à la Oneida Jean-Pierre Koenig and Karin Michelson Gender–number marking in Archi: Small is complex Marina Chumakina and Greville G. Corbett Part III. Measuring Complexity Contrasting modes of representation for inflectional systems: Some implications for computing morphological complexity Gregory Stump and Raphael A. Finkel Computational complexity of abstractive morphology Vito Pirrelli, Marcello Ferro, and Claudia Marzi Patterns of syncretism and paradigm complexity: The case of Old and Middle Indic declension Paolo Milizia OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi vi Contents Learning and the complexity of Ø-marking Sebastian Bank and Jochen Trommer References Languages Index Names Index Subject Index OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi List of Figures and Tables Figures Figure 3.1 Kayardild clause structure and attachment of features 31 Figure 3.2 Features which percolate onto word ω 34 Figure 5.1 A hierarchy of nom-index 80 Figure 5.2 A hierarchy of person values 80 Figure 5.3 A hierarchy of gender values 80 Figure 5.4 A hierarchy of number values 80 Figure 6.1 Types of simple dynamic verbs according to the phonological shape of their perfective stem (total 142) 112 Figure 8.1 One-to-one and one-to-many relations in one-node-per-variable and many-nodes-per-variable neural networks 146 Figure 8.2 A two-level finite state transducer for the Italian irregular form vengono ‘they come’ 150 Figure 8.3 Outline architecture of a TSOM and a two-dimensional 20×20 TSOM 152 Figure 8.4 Topological propagation of long-term potentiation and long-term depression of temporal re-entrant connections over two successive time steps; and word-graph representation of German past participles 154 Figure 8.5 Topological dispersion of symbols on temporal and spatio-temporal maps, plotted by their position in input words 157 Figure 8.6 Alignment plots of the finden paradigm on a temporal (left) and a spatio-temporal (right) map 158 Figure 8.7 (Topological) dispersion values across activation patterns triggered by selected inflectional endings on temporal and spatio-temporal maps for Italian and German known word forms,andonunknownwordforms 159 Figure 8.8 BMU activation chains for vediamo-vedete-crediamo on a 20×20 map (left) and their word-graph representation (right) 160 Figure 8.9 Correlation coefficients between alignment scores and recall scores for novel words, plotted by letter positions relative to the stem-ending boundary 161 OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi viii List of Figures and Tables Figure 8.10 Stem dispersion over <PREF-X, X > pairs in Italian and German verb forms. Dispersion values are given as a fraction of 100overthemap’sdiagonal 163 Figure 9.1 A graph representing the relative frequencies of morphosyntactic property combinations in Old Indic adjectival paradigms 172 Tables Table 3.1 Morphome types and their mode of categorization 29 Table 3.2 Inflectional features and their potential values 32 Table 3.3 Features for which lexical stems can inflect 37 Table 3.4 Inflectional features and their morphomic exponents 51 Table 3.5 Phonological exponents of morphomes 51 Table 4.1 Inflection for ampl ‘laugh at’ with different subjects, objects, and tenses 56 Table 4.2 Free pronouns in Kanum, absolutive, and ergative forms 56 Table4.3 Kanumverbalinflection 57 Table 4.4 Inflection for makr ‘roast’ with different subjects, objects, and tenses 61 Table 4.5 Opacity and transparency in pronominal systems 62 Table 4.6 Tense distinctions portmanteau with agreement affixes 62 Table 4.7 Pronominal distinctions in the idealized agreement affixes 63 Table 4.8 Takeovers in the object prefixes for two verbs 63 Table 4.9 Referrals in the object prefixes 64 Table 4.10 Referrals in tense affixes 64 Table 4.11 Referrals in the subject suffixes 64 Table 5.1 Oneida prenominal prefixesC ( -stem allomorphs) 73 Table 5.2 The nineteen possible categories of semantic indices 79 Table 5.3 Number of morphs in each stem class from which all other fifty-seven forms can be deduced 88 Table 5.4 Number and percentage of stems of each class in twelve texts 89 Table 6.1 Agreeing lexical items in the Archi dictionary 95 Table 6.2 Syncretism pattern A 98 Table 6.3 Syncretism pattern B 98 Table 6.4 Archi prefixes 98 Table 6.5 Archi infixes, Set I 98 OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi List of Figures and Tables ix Table 6.6 Archi infixes, Set II 99 Table 6.7 Archi suffixes 99 Table 6.8 Partial paradigm of first and second person pronouns 102 Table 6.9 Verbs of similar phonology, but different patterns of realizing agreement 111 Table 7.1 Orthographic forms of the verbs in (1) 121 Table 7.2 Hearer-oriented plat for the verbs in (1) 122 Table 7.3 Speaker-oriented plat for the verbs in (1) 123 Table 7.4 Morphological indices employed in the speaker-oriented plat 124 Table 7.5 Two hypothetical plats 125 Table 7.6 Plat representing a hypothetical system of ICs 127 Table 7.7 Inflection classes, exponences, and distillations in the two plats 128 Table 7.8 4-MPS entropy (× 100) of the four distillations in Table 7.7 129 Table 7.9 Distinct exponences in each of the four distillations 129 Table 7.10 Traditional principal parts of five Latin verbs 130 Table 7.11 Optimal dynamic principal-part sets of three verbs (H-O plat) 131 Table 7.12 Dynamic principal-part sets in the two plats 132 Table 7.13 Dynamic principal-part numbers in the two plats 132 Table 7.14 Candidate principal-part sets for cast in the H-O plat 133 Table 7.15 Candidate principal-part sets for cast (S-O plat) 133 Table 7.16 Number of viable optimal dynamic principal-part sets 134 Table 7.17 Candidate principal-part sets for pass (H-O plat) 134 Table 7.18 Candidate principal-part sets for pass (S-O plat) 135 Table 7.19 Density of viable dynamic principal-part sets among all candidate sets having the same number of members 135 Table 7.20 IC predictability of twelve verbs in the H-O and S-O plats 137 Table 7.21 Cell predictability measures for thirteen verbs in the H-O and S-O plats 138 Table 7.22 Predictiveness of a verb’s cells, averaged across verbs 139 Table 7.23 Predictiveness of a verb’s past-tense cell in the H-O and S-O plats 139 Table 9.1 Relative frequencies of inflectional values on the basis of Lanman (1880) 171 Table 9.2 Old Indic -a-/-a¯- adjective declension and relative frequencies of the sets of inflectional value arrays associated with the different exponents 173 OUP CORRECTED PROOF – FINAL, 5/3/2015, SPi x List of Figures and Tables Table 9.3 R and D values for some hypothetical variants of the paradigm in Table 9.2 174 Table 9.4 Pali -a-/-a-¯ adjective declension 175 Table 9.5 Syncretism in the marked gender and in the marked number in Pali -a-/-a-¯ adjectives 176 Table 9.6 Old Church Slavic (a) and Russian (b) definite adjective 178 Table 9.7 Gender syncretism and gender semi-separate exponence in the plural of the Latin -o-/-a¯ and of the Pali -a-/-a-¯ adjectives 179 Table 9.8 Vertical and