Statistical modeling of language universals

Synthese, 2006

In this article we discuss the notion of a linguistic universal, and possible sources of such invariant properties of natural languages. In the first part, we explore the conceptual issues that arise. In the second part of the paper, we focus on the explanatory potential of horizontal evolution. We particularly focus on two case studies, concerning Zipf's Law and universal properties of color terms, respectively. We show how computer simulations can be employed to study the large scale, emergent, consequences of psychologically and psychologically motivated assumptions about the working of horizontal language transmission. Keywords Language universals • Evolution • Game theory 1 Language universals There are around 5,000 spoken languages of the world today, and they are all different. Thus, natural languages are very diverse. However, most linguists share the view that the languages of the world, and the way they are used in different communities, have a lot in common. The properties that all languages have in common are called language universals. Linguists have claimed of many different (kinds of) things that they are language universals. For a simple example of a syntactic universal, it has been claimed that all languages spoken by humans have, and make a distinction between, nouns,

Dunn et al. (2011) use a dynamic model of word order correlations in four phylogenies (language families) to argue that Greenbergian word order correlations are lineage-specific rather than universal. Dunn et al.’s model represents an important advance for diachronic typology (specifically, the dynamicization of a synchronic typology). However, certain assumptions made by Dunn and colleagues in the application of the model pose serious issues in accepting the conclusions, notably the absence of any Type II error analysis to assess the rate of false negatives, the absence of contact effects and the nature of the phylogenies used. Nevertheless, typologists should welcome the model and encourage the development of a revised model with more linguistically plausible assumptions.

Journal of the Royal Society, Interface / the Royal Society, 2014

The frequency with which we use different words changes all the time, and every so often, a new lexical item is invented or another one ceases to be used. Beyond a small sample of lexical items whose properties are well studied, little is known about the dynamics of lexical evolution. How do the lexical inventories of languages, viewed as entire systems, evolve? Is the rate of evolution of the lexicon contingent upon historical factors or is it driven by regularities, perhaps to do with universals of cognition and social interaction? We address these questions using the Google Books N-Gram Corpus as a source of data and relative entropy as a measure of changes in the frequency distributions of words. It turns out that there are both universals and historical contingencies at work. Across several languages, we observe similar rates of change, but only at timescales of at least around five decades. At shorter timescales, the rate of change is highly variable and differs between langua...

Journal of English Linguistics, 2001

The vast literature on the reasons why human languages inevitably change through time focuses on two types of causation, "functional" and "social." My purpose here is to explore a third category of explanation, one that has largely been ignored or dismissed as seemingly inconsequential: I argue that a chance/chaos model of linguistic change is a necessary and important supplement to functional and social explanations.

2014

The frequency with which we use different words changes all the time, and every so often, a new lexical item is invented or another one ceases to be used. Beyond a small sample of lexical items whose properties are well studied, little is known about the dynamics of lexical evolution. How do the lexical inventories of languages, viewed as entire systems, evolve? Is the rate of evolution of the lexicon contingent upon historical factors or is it driven by regularities, perhaps to do with universals of cognition and social interaction? We address these questions using the Google Books N-Gram Corpus as a source of data and relative entropy as a measure of changes in the frequency distributions of words. It turns out that there are both universals and historical contingencies at work. Across several languages, we observe similar rates of change, but only at timescales of at least around five decades. At shorter timescales, the rate of change is highly variable and differs between languages. Major societal transformations as well as catastrophic events such as wars lead to increased change in frequency distributions, whereas stability in society has a dampening effect on lexical evolution.

Rosemeyer, Malte (In press): Modelling frequency effects in language change. In Behrens, Heike and Stefan Pfänder (eds.), Again on Frequency. Effects in Language. Berlin: Mouton de Gruyter., 2014

Processes of language change in which a grammatical construction decreases in usage frequency should be modeled in terms of both type and token frequency. This paper analyzes Spanish compound tense auxiliary selection, suggesting that the replacement of ser 'be' with haber 'have' was affected by (a) the salience of haber + participle in usage contexts previously associated with the use of ser + participle, and (b) the general token frequency of specific verbs and of the ser + participle syntagms that form from these verbs. I argue that it is necessary to account for both historical processes in order to explain the synchronic gradience in auxiliary selection posited in influential Auxiliary Selection Hierarchy (ASH), and propose statistical methodology to model these frequency effects in language change. The findings suggest both type and token frequency effects. The former is an actualization process due to the prototypicality of use of ser + participle with telic predicates implying a change of state (e.g. morir 'die'), which are affected by the replacement with haber + participle at a later point in the process of change, whereas the latter is a conservation process evident in ser + participle syntagms formed from highly frequent verbs that, due to their high token frequency, are less affected by the ongoing change. This skewed frequency distribution in the verb population and the resulting conserving effect leads to further changes in auxiliary selection rules.

Canadian Journal of Linguistics-revue Canadienne De Linguistique, 2010

Linguistic Typology at the Crossroads, 2023

We examine a database of 3089 languages coded for 351 morphosyntactic features, including almost all of the morphosyntactic features found in The World Atlas of Language Structures (Dryer & Haspelmath 2013). We apply Factor Analysis of Mixed Data, and determine that the main dimensions of global morphological variation involve (1) word order in clauses and adpositional phrases, (2) head-versus dependent-marking, and (3) a set of features that show an east-west distribution. We find roughly the same features clustering in similar dimensions when we examine individual macro-areas, thus confirming the universal relevance of these groupings of features, as encapsulated in well-known implicational universals. This study confirms established insights in linguistic typology, extending earlier research to a much larger set of languages, and uncovers a number of areal patterns in the data.

Nature, 2011

2012

Certain grammatical patterns are found again and again in the languages of the world. Some of these patterns recur so frequently that they are given the label “universal”. Explaining the source of such patterns is clearly an important goal of linguistics, but how to go about doing this is not obvious. Problems range from the terminological (what sort of patterns should we consider universal?) to the methodological (what kind of explanation will we accept as sufficient?) to the theoretical (what role does a universal grammar have in shaping recurrent patterns? what role do functional considerations play?). How one answers one of these questions will affect how one answers the others. Can probabilistic generalizations be considered universals? If so, then we need explanations predicting probabilistic patterns. Are we looking for proximate explanations (for example, “language A shows pattern X because it inherited it from its parent language”) or ultimate ones (for example, “language A shows pattern X because only this pattern is permitted by Universal Grammar”)? Will we assume there is no such thing as Universal Grammar? Then, of course, we cannot appeal to it for any sort of explanation. Will we assume there is such a thing? Then, what is its precise structure?

Wiley Interdisciplinary Reviews: Cognitive Science, 2014

Linguistic typological preferences have often been linked to cognitive processing preferences but often without recourse to typologically relevant experiments on cognitive processing. This article reviews experimental work on the possible parallels between preferences in cognitive processing and language typology. I summarize the main theoretical accounts of the processing-typology connection and show that typological distributions arise diachronically from preferred paths of language change, which may be affected by the degree to which alternative structures are preferred (e.g., easier) in acquisition or usage. The surveyed experimental evidence shows that considerable support exists for many linguistic universals to reflect preferences in cognitive processing. Artificial language learning experiments emerge as a promising method for researching the processing-typology connection, as long as its limitations are taken into account. I further show that social and cultural differences in cognition may have an effect on typological distributions and that to account for this variation a multidisciplinary approach to the processing-typology connection has to be developed. Lastly, since the body of experimental research does not adequately represent the linguistic diversity of the world's languages, it remains as an urgent task for the field to better account for this diversity in future work. WIREs Cogn Sci 2014, 5:477-487. doi: 10.1002/wcs.1294 For further resources related to this article, please visit the WIREs website. The author has declared no conflicts of interest for this article.

Cette thèse se propose d'étudier la grammaticalisation, processus d'évolution linguistique par lequel les éléments fonctionnels de la langue se trouvent remplacés au cours du temps par des mots ou des constructions de contenu, c'est-à-dire servant à désigner des entités plus concrètes. La grammaticalisation est donc un cas particulier de remplacement sémantique. Or, la langue faisant l'objet d'un consensus social bien établi, il semble que le changement sémantique s'effectue à contre-courant de la bonne efficacité de la communication ; pourtant, il est attesté dans toutes les langues, toutes les époques et, comme le montre la grammaticalisation, toutes les catégories linguistiques. Dans cette thèse, nous étudions d'abord le phénomène de grammaticalisation d'un point de vue empirique, en analysant les fréquences d'usage de plusieurs centaines de constructions du langage connaissant une ou plusieurs grammaticalisations au cours de l'histoire de la langue française. Ces profils de fréquence sont extraits de la base de données de Frantext, qui permet de couvrir une période de sept siècles. L'augmentation de fréquence en courbe en S concomitante du remplacement sémantique, attestée dans la littérature, est confirmée, mais aussi complétée par l'observation d'une période de latence, une stagnation de la fréquence d'usage de la construction alors même que celle-ci manifeste déjà son nouveau sens. Les distributions statistiques des observables décrivant ces deux phénomènes sont obtenues et quantifiées. Un modèle de marche aléatoire est ensuite proposé reproduisant ces deux phénomènes. La latence s'y trouve expliquée comme un phénomène critique, au voisinage d'une bifurcation point-col. Une extension de ce modèle articulant l'organisation du réseau sémantique et les formes possibles de l'évolution est ensuite discutée.

Language, 2000

NOTICES fragments of Japanese (233-68) and English (269-325). Finally, it should be noted that C does not compare DP with the current dominant theory in generative phonology, optimality theory (OT). In part, this is because OT is a theory of constraint combination, not phonological representation, and thus OT can benefit from the theoretical advances made by C. Conversely, much of the early literature demonstrating the superiority of OT to derivational approaches does not reflect on DP, either. The primary difference between theories is the violability of constraints in OT. Since DP is the more restrictive theory, the burden of proof is on practitioners of OT, who would greatly benefit from a careful reading of this exceptional volume. [STEFAN FRISCH, University of Michigan.] Linguistic structure and linguistic change: Explanation from language processing. By THOMAS BERG. Oxford & New York: Clarendon Press, 1998. Pp. xiii, 336. The objective of this formidable book is to establish a relationship between psycholinguistic processing and linguistic patterns. Berg's approach is to derive predictions from psycholinguistic research and to test them against an impressively wide array of linguistic patterns and diachronic processes as well as those found in poetics. B's case is compelling, though he acknowledges that a successful outcome means that the psycholinguistic approach is merely plausible as it is impossible to eliminate all alternative analyses (65). Although B admits that language is a multifaceted phenomenon that is real at the neurological, physical, sociological, and psychological levels, he adopts a psycholinguistic approach because it bears on all of these levels and because it is a priori feasible given that speaking and hearing are psychological activities so fundamental that productive and perceptual processes are likely to exert influence on the information to be processed, namely, language (54); clearly, B believes that language is NOT autonomous, contrary to many generativists' views. Further, B cautions against eclecticism, as principles that are adopted from one framework are couched in larger, coherent theories, and their explanatory force is eviscerated when that context disappears. B heeds his own advice and seeks to determine how far a single model can take us in explaining language. B's answer is, quite far, although when the relevant disciplines advance enough to be able to assess their contributions, B conjectures that linguistic patterns will undoubtedly be understood as the result of multicausal efforts (279). In Ch. 1, 'On the "art" of explanation' (1-17), B discusses explanation in scientific theory and in fragments of Japanese (233-68) and English (269-325). Finally, it should be noted that C does not compare DP with the current dominant theory in generative phonology, optimality theory (OT). In part, this is because OT is a theory of constraint combination, not phonological representation, and thus OT can benefit from the theoretical advances made by C. Conversely, much of the early literature demonstrating the superiority of OT to derivational approaches does not reflect on DP, either. The primary difference between theories is the violability of constraints in OT. Since DP is the more restrictive theory, the burden of proof is on practitioners of OT, who would greatly benefit from a careful reading of this exceptional volume. [STEFAN FRISCH, University of Michigan.] Linguistic structure and linguistic change: Explanation from language processing. By THOMAS BERG. Oxford & New York: Clarendon Press, 1998. Pp. xiii, 336. The objective of this formidable book is to establish a relationship between psycholinguistic processing and linguistic patterns. Berg's approach is to derive predictions from psycholinguistic research and to test them against an impressively wide array of linguistic patterns and diachronic processes as well as those found in poetics. B's case is compelling, though he acknowledges that a successful outcome means that the psycholinguistic approach is merely plausible as it is impossible to eliminate all alternative analyses (65). Although B admits that language is a multifaceted phenomenon that is real at the neurological, physical, sociological, and psychological levels, he adopts a psycholinguistic approach because it bears on all of these levels and because it is a priori feasible given that speaking and hearing are psychological activities so fundamental that productive and perceptual processes are likely to exert influence on the information to be processed, namely, language (54); clearly, B believes that language is NOT autonomous, contrary to many generativists' views. Further, B cautions against eclecticism, as principles that are adopted from one framework are couched in larger, coherent theories, and their explanatory force is eviscerated when that context disappears. B heeds his own advice and seeks to determine how far a single model can take us in explaining language. B's answer is, quite far, although when the relevant disciplines advance enough to be able to assess their contributions, B conjectures that linguistic patterns will undoubtedly be understood as the result of multicausal efforts (279). In Ch. 1, 'On the "art" of explanation' (1-17), B discusses explanation in scientific theory and in fragments of Japanese (233-68) and English (269-325). Finally, it should be noted that C does not compare DP with the current dominant theory in generative phonology, optimality theory (OT). In part, this is because OT is a theory of constraint combination, not phonological representation, and thus OT can benefit from the theoretical advances made by C. Conversely, much of the early literature demonstrating the superiority of OT to derivational approaches does not reflect on DP, either. The primary difference between theories is the violability of constraints in OT. Since DP is the more restrictive theory, the burden of proof is on practitioners of OT, who would greatly benefit from a careful reading of this exceptional volume. [STEFAN FRISCH, University of Michigan.] Linguistic structure and linguistic change: Explanation from language processing. By THOMAS BERG. Oxford & New York: Clarendon Press, 1998. Pp. xiii, 336. The objective of this formidable book is to establish a relationship between psycholinguistic processing and linguistic patterns. Berg's approach is to derive predictions from psycholinguistic research and to test them against an impressively wide array of linguistic patterns and diachronic processes as well as those found in poetics. B's case is compelling, though he acknowledges that a successful outcome means that the psycholinguistic approach is merely plausible as it is impossible to eliminate all alternative analyses (65). Although B admits that language is a multifaceted phenomenon that is real at the neurological, physical, sociological, and psychological levels, he adopts a psycholinguistic approach because it bears on all of these levels and because it is a priori feasible given that speaking and hearing are psychological activities so fundamental that productive and perceptual processes are likely to exert influence on the information to be processed, namely, language (54); clearly, B believes that language is NOT autonomous, contrary to many generativists' views. Further, B cautions against eclecticism, as principles that are adopted from one framework are couched in larger, coherent theories, and their explanatory force is eviscerated when that context disappears. B heeds his own advice and seeks to determine how far a single model can take us in explaining language. B's answer is, quite far, although when the relevant disciplines advance enough to be able to assess their contributions, B conjectures that linguistic patterns will undoubtedly be understood as the result of multicausal efforts (279). In Ch. 1, 'On the "art" of explanation' (1-17), B discusses explanation in scientific theory and in fragments of Japanese (233-68) and English (269-325). Finally, it should be noted that C does not compare DP with the current dominant theory in generative phonology, optimality theory (OT). In part, this is because OT is a theory of constraint combination, not phonological representation, and thus OT can benefit from the theoretical advances made by C. Conversely, much of the early literature demonstrating the superiority of OT to derivational approaches does not reflect on DP, either. The primary difference between theories is the violability of constraints in OT. Since DP is the more restrictive theory, the burden of proof is on practitioners of OT, who would greatly benefit from a careful reading of this exceptional volume. [STEFAN FRISCH, University of Michigan.] Linguistic structure and linguistic change: Explanation from language processing. By THOMAS BERG. Oxford & New York: Clarendon Press, 1998. Pp. xiii, 336. The objective of this formidable book is to establish a relationship between psycholinguistic processing and linguistic patterns. Berg's approach is to derive predictions from psycholinguistic research and to test them against an impressively wide array of linguistic patterns and diachronic processes as well as those found in poetics. B's case is compelling, though he acknowledges that a successful outcome means that the psycholinguistic approach is merely plausible as it is impossible to eliminate all alternative analyses (65). Although B admits that language is a multifaceted phenomenon that is real at the neurological, physical, sociological, and psychological levels, he adopts a psycholinguistic approach because it bears on all of these levels and because it is a priori feasible given that speaking and hearing are psychological activities so fundamental that productive and perceptual processes are likely to exert influence on the information to be processed, namely, language (54); clearly, B believes that language is...