Language Evolution and Computation Bibliography

How does the behavior of individuals in a society influence whether knowledge accumulates over generations? We explore this question using a simple model of cultural evolution as a process of ``iterated learning,'' where each agent in a sequence learns and passes on a piece of information. Using both mathematical analyses involving rational Bayesian agents and laboratory experiments with human participants, we vary whether agents observe data from the environment and what kind of information they receive from the previous agent. Our mathematical and empirical results both suggest that merely observing the behavior of other learners is not sufficient to produce cumulative cultural evolution, but that knowledge can be accumulated over generations when agents are able to communicate the plausibility of different hypotheses.

The past couple of decades have seen an explosion of research on language evolution, initially fueled by Pinker and Bloomas (1990) groundbreaking article arguing for the natural selection of biological structures dedicated to language. The new millennium has seen a shift toward explaining language evolution in terms of cultural evolution rather than biological adaptation. Crucially, this research has many important implications for cognitive science, not only in terms of the nature of the biases to consider in language acquisition but also for cognition, more generally. In this symposium, we therefore take stock of current work on the cultural evolution of language, highlighting key implications of this work for cognitive scientists from different perspectives, ranging from philosophical considerations (Chater) and Bayesian analyses (Griffiths) to evolutionary psycholinguistics (Kirby) and molecular genetics (Christiansen).

We develop a Bayesian Iterated Learning Model (BILM) that models the cultural evolution of language as it is transmitted over generations of learners. We study the outcome of iterated learning in relation to the behavior of individual agents (their biases) and the social structure through which they transmit their behavior. BILM makes individual learning biases explicit and offers a direct comparison of how individual biases relate to the outcome of iterated learning. Most earlier BILMs use simple one parent to one child (monadic) chains of homogeneous learners to study the outcome of iterated learning in terms of bias manipulations. Here, we develop a BILM to study two novel manipulations in social parameters: population size and population heterogeneity, to determine more precisely what the transmission process itself can add to the outcome of iterated learning. Our monadic model replicates the existing BILM results, however our manipulations show that the outcome of iterated learning is sensitive to more factors than are explicitly encoded in the prior. This calls into question the appropriateness of assuming strong Bayesian inference in the iterated learning framework and has important implications for the study of language evolution in general.

Cognitive scientists tend to focus on the behavior of single individuals thinking and perceiving on their own. This is natural because our own introspection provides us with unique insight into this level. However, interacting groups of people also create emergent structures that are not intentionally produced by any individual. People participate in collective behavior patterns that they may not even be able to perceive, let alone understand. Social phenomena such as rumors, linguistic conventions, the emergence of a standard currency, transportation systems, the World Wide Web, resource harvesting, crowding, and scientific establishments arise because of individualsa beliefs and goals, but the eventual form that these phenomena take is rarely the goal of any individual.

Similarities between human languages are often taken as evidence of constraints on language learning. However, such similarities could also be the result of descent from a common ancestor. In the framework of iterated learning, language evolution converges to an equilibrium that is independent of its starting point, with the effect of shared ancestry decaying over time. Therefore, the central question is the rate of this convergence, which we formally analyze here. We show that convergence occurs in a number of generations that is O(n log n) for Bayesian learning of the ranking of n constraints or the values of n binary parameters. We also present simulations confirming this result and indicating how convergence is affected by the entropy of the prior distribution over languages.

In his book Grooming, Gossip and the Evolution of Language, biologist Robin Dunbar (1997) proposes a new way of looking at the evolution of language. According to this view, language evolved to provide a new social bonding mechanism: Gossiping. This allows humans to live in larger groups than other primates, which increasing predation risks forced our ancestors to do. We use a computational multi-agent model to test the internal workings of this hypothesis, with interesting results. Our work provides a fundamentally new kind of evidence for Dunbaras theory, by experimentally demonstrating that greater group sizes can stimulate the evolution of language as a tool for social cohesion.

Existing accounts of the origin of human communication assume a pre-existing behavioral system shared among members of a social group. This paper is concerned with the origin of that system; specifically, it explores its characteristics and functionality as well as the circumstances under which it could have appeared. A number of agent-based computer simulations test whether the capacities for arbitrary imitation and pattern completion can lead to a behavioral system that could be co-opted for communication. The results show that arbitrary imitation and pattern completion may indeed generate a population-wide shared behavioral system whose structure reflects the structure of the environment, and therefore could easily have been co-opted for communication. This system may have paved the way for other biological capacities widely believed to be necessary for communication, such as shared intentionality and symbolicity, to co-evolve.

Human languages include vast numbers of learned, arbitrary signal-meaning mappings but also many complex signal-meaning mappings that are systematically related to each other (i.e. not arbitrary). Although arbitrariness and systematicity are clearly related, the development of the two in communication systems has been explored independently. We present an experiment in which participants invent signs from scratch to refer to a set of real concepts that share semantic features. Through interaction, the systematic re-use of arbitrary elements emerges.

Imitation alone cannot explain language evolution. Two additional ingredients have been proposed that may help explain the evolution of language systems: learning biases and social collaboration. An experimental method was developed that isolated the roles of collaboration and learning biases in the development of novel communication systems. Participants played a Pictionary-like task to develop ad hoc graphical communication systems in one of two conditions: one in which they interacted with a partner (Interaction condition), and one in which they received the same images from a apseudo-partnera but did not interact (Pseudo-Interaction condition). Comparison of the resultant communication systems showed that the Interaction condition yielded higher identification accuracy, greater refinement of graphical signs and more alignment on a set of shared graphical signs (in fact, graphical alignment did not occur at all in the Pseudo-Interaction condition). Thus, collaboration plays a crucial role in the evolution of human communication systems.

In this paper, we attempt to explain the emergence of the linguistic diversity that exists across the consonant inventories of some of the major language families of the world through a complex network based growth model. There is only a single parameter for this model that is meant to introduce a small amount of randomness in the otherwise preferential attachment based growth process. The experiments with this model parameter indicates that the choice of consonants among the languages within a family are far more preferential than it is across the families. The implications of this result are twofold -- (a) there is an innate preference of the speakers towards acquiring certain linguistic structures over others and (b) shared ancestry propels the stronger preferential connection between the languages within a family than across them. Furthermore, our observations indicate that this parameter might bear a correlation with the period of existence of the language families under investigation.

A key feature of many biological distributed systems is that they have the capacity to behave in highly coordinated ways. In the domain of language, such coordination dynamics have been studied within the framework of language games. As yet however, a fundamental understanding that goes beyond the simplest cases is still missing. In this paper, a novel approach is proposed for investigating coordination problems. I illustrate the approach for a simple but well studied case called the naming game. I will therefore bring together a number of ideas from Artificial Chemistry and Chemical Reaction Network Theory, Semiotic Dynamics and Immunology, and conclude by arguing why the proposed approach provides a good starting point for tackling more complex coordination problems as well.

We conduct an evolutionary simulation to explore the coevolution of language and a language-related ability, intentionality sharing. Our simulation shows that during the evolution of a simple informative language, communicative success helps optimize the level of intentionality sharing in the population. This study illustrates a selective role of language communications on language-related abilities, and assists the discussion of the uniqueness of language-related abilities based on comparative studies.

Recent research has shown that language and the socio-cognitive phenomena associated with it can be aptly modeled and visualized through networks of linguistic entities. However, most of the existing works on linguistic networks focus only on the local properties of the networks. This study is an attempt to analyze the structure of languages via a purely structural technique, namely spectral analysis, which is ideally suited for discovering the global correlations in a network. Application of this technique to PhoNet, the co-occurrence network of consonants, not only reveals several natural linguistic principles governing the structure of the consonant inventories, but is also able to quantify their relative importance. We believe that this powerful technique can be successfully applied, in general, to study the structure of natural languages.

Creole is a new born language emerging in most cases where language contact takes place. Simulating behaviors that creole communities are formed in some environments, we could contribute to actual proof of some linguistic theories concerning language acquisition. Thus far, a simulation study of the emergence of creoles has been reported in the mathematical framework. In this paper we introduce a spatial structure to the framework. We show that local creole communities are organized, and creolization may occur when language learners learn of ten from non-parental language speakers, in contrast to the non-spatial model. The quantitative analysis of the result tells us that emergence of local colonies at the early stage tends to induce the full creolization.

We simulate the evolution of a domain language in small speaker communities. Data from experiments (Garrod et al., 2007; Fay et al., 2008) show that human communicators can evolve graphical languages quickly in a constrained task (Pictionary), and that communities converge towards a common language even in the absence of feedback about the success of each communication. We postulate that simulations of such horizontal evolution have to take into account properties of human memory (cue-based retrieval, learning, decay). We implement a model that can draw abstract concepts through sets of non-abstract, related concepts, and recognize such drawings. The knowledge base is a network with association strengths randomly sampled from a natural distribution found in a text corpus; it is a mixture of knowledge shared between agents and individual knowledge. In three experiments, we show that the agent communities converge, but that initial convergence is stronger when communities are structured so that the same pairs of agents interact throughout. Convergence is weaker in communities when agents do not swap roles (between recognizing and drawing), predicting the necessity of bi-directional communication in domain language evolution. Average and ultimate recognition performance depends on how much of the knowledge agents share initially.

This paper presents a new statistical methodfor detecting and tracking changes in word meaning, based on Latent Semantic Analysis. By comparing the density of semantic vector clusters this method allows researchers to make statistical inferences on questions such as whether the meaning of a word changed across time or if a phonetic cluster is associated with a specific meaning. Possible applications of this method are then illustrated in tracing the semantic change of `dog', `do', and `deer' in early English and examining and comparing phonaesthemes.

Culture is typically viewed as consisting of traits inherited epigenetically, through social learning. However, cultural diversity has species-typical constraints(1), presumably of genetic origin. A celebrated, if contentious, example is whether a universal grammar constrains syntactic diversity in human languages(2). Oscine songbirds exhibit song learning and provide biologically tractable models of culture: members of a species show individual variation in song(3) and geographically separated groups have local song dialects(4,5). Different species exhibit distinct song cultures(6,7), suggestive of genetic constraints(8,9). Without such constraints, innovations and copying errors should cause unbounded variation over multiple generations or geographical distance, contrary to observations(9). Here we report an experiment designed to determine whether wild-type song culture might emerge over multiple generations in an isolated colony founded by isolates, and, if so, how this might happen and what type of social environment is required(10). Zebra finch isolates, unexposed to singing males during development, produce song with characteristics that differ from the wild-type song found in laboratory(11) or natural colonies. In tutoring lineages starting from isolate founders, we quantified alterations in song across tutoring generations in two social environments: tutor-pupil pairs in sound-isolated chambers and an isolated semi-natural colony. In both settings, juveniles imitated the isolate tutors but changed certain characteristics of the songs. These alterations accumulated over learning generations. Consequently, songs evolved towards the wild-type in three to four generations. Thus, species-typical song culture can appear de novo. Our study has parallels with language change and evolution(12-14). In analogy to models in quantitative genetics(15,16), we model song culture as a multigenerational phenotype partly encoded genetically in an isolate founding population, influenced by environmental variables and taking multiple generations to emerge.

The signalling pathways controlling both the evolution and development of language in the human brain remain unknown. So far, the transcription factor FOXP2 (forkhead box P2) is the only gene implicated in Mendelian forms of human speech and language dysfunction. It has been proposed that the amino acid composition in the human variant of FOXP2 has undergone accelerated evolution, and this two-amino-acid change occurred around the time of language emergence in humans. However, this remains controversial, and whether the acquisition of these amino acids in human FOXP2 has any functional consequence in human neurons remains untested. Here we demonstrate that these two human-specific amino acids alter FOXP2 function by conferring differential transcriptional regulation in vitro. We extend these observations in vivo to human and chimpanzee brain, and use network analysis to identify novel relationships among the differentially expressed genes. These data provide experimental support for the functional relevance of changes in FOXP2 that occur on the human lineage, highlighting specific pathways with direct consequences for human brain development and disease in the central nervous system (CNS). Because FOXP2 has an important role in speech and language in humans, the identified targets may have a critical function in the development and evolution of language circuitry in humans.

Debates about human prehistory often center on the role that population expansions play in shaping biological and cultural diversity. Hypotheses on the origin of the Austronesian settlers of the Pacific are divided between a recent 'pulse-pause' expansion from Taiwan and an older 'slow-boat' diffusion from Wallacea. We used lexical data and Bayesian phylogenetic methods to construct a phylogeny of 400 languages. In agreement with the pulse-pause scenario, the language trees place the Austronesian origin in Taiwan approximately 5230 years ago and reveal a series of settlement pauses and expansion pulses linked to technological and social innovations. These results are robust to assumptions about the rooting and calibration of the trees and demonstrate the combined power of linguistic scholarship, database technologies, and computational phylogenetic methods for resolving questions about human prehistory.

Two prehistoric migrations peopled the Pacific. One reached New Guinea and Australia, and a second, more recent, migration extended through Melanesia and from there to the Polynesian islands. These migrations were accompanied by two distinct populations of the specific human pathogen Helicobacter pylori, called hpSahul and hspMaori, respectively. hpSahul split from Asian populations of H. pylori 31,000 to 37,000 years ago, in concordance with archaeological history. The hpSahul populations in New Guinea and Australia have diverged sufficiently to indicate that they have remained isolated for the past 23,000 to 32,000 years. The second human expansion from Taiwan 5000 years ago dispersed one of several subgroups of the Austronesian language family along with one of several hspMaori clades into Melanesia and Polynesia, where both language and parasite have continued to diverge.

The enormous increase of popularity and use of the worldwide web has led in the recent years to important changes in the ways people communicate. An interesting example of this fact is provided by the now very popular social annotation systems, through which users annotate resources (such as web pages or digital photographs) with keywords known as ``tags.'' Understanding the rich emergent structures resulting from the uncoordinated actions of users calls for an interdisciplinary effort. In particular concepts borrowed from statistical physics, such as random walks (RWs), and complex networks theory, can effectively contribute to the mathematical modeling of social annotation systems. Here, we show that the process of social annotation can be seen as a collective but uncoordinated exploration of an underlying semantic space, pictured as a graph, through a series of RWs. This modeling framework reproduces several aspects, thus far unexplained, of social annotation, among which are the peculiar growth of the size of the vocabulary used by the community and its complex network structure that represents an externalization of semantic structures grounded in cognition and that are typically hard to access.

Language acquisition and processing are governed by genetic constraints. A crucial unresolved question is how far these genetic constraints have coevolved with language, perhaps resulting in a highly specialized and species-specific language 'module,' and how much language acquisition and processing redeploy preexisting cognitive machinery. In the present work, we explored the circumstances under which genes encoding language-specific properties could have coevolved with language itself. We present a theoretical model, implemented in computer simulations, of key aspects of the interaction of genes and language. Our results show that genes for language could have coevolved only with highly stable aspects of the linguistic environment; a rapidly changing linguistic environment does not provide a stable target for natural selection. Thus, a biological endowment could not coevolve with properties of language that began as learned cultural conventions, because cultural conventions change much more rapidly than genes. We argue that this rules out the possibility that arbitrary properties of language, including abstract syntactic principles governing phrase structure, case marking, and agreement, have been built into a 'language module' by natural selection. The genetic basis of human language acquisition and processing did not coevolve with language, but primarily predates the emergence of language. As suggested by Darwin, the fit between language and its underlying mechanisms arose because language has evolved to fit the human brain, rather than the reverse.

Language acquisition maps linguistic experience, primary linguistic data (PLD), onto linguistic knowledge, a grammar. Classically, computational models of language acquisition assume a single target grammar and one PLD source, the central question being whether the target grammar can be acquired from the PLD. However, real-world learners confront populations with variation, i.e., multiple target grammars and PLDs. Removing this idealization has inspired a new class of population-based language acquisition models. This paper contrasts 2 such models. In the first, iterated learning (IL), each learner receives PLD from one target grammar but different learners can have different targets. In the second, social learning (SL), each learner receives PLD from possibly multiple targets, e. g., from 2 parents. We demonstrate that these 2 models have radically different evolutionary consequences. The IL model is dynamically deficient in 2 key respects. First, the IL model admits only linear dynamics and so cannot describe phase transitions, attested rapid changes in languages over time. Second, the IL model cannot properly describe the stability of languages over time. In contrast, the SL model leads to nonlinear dynamics, bifurcations, and possibly multiple equilibria and so suffices to model both the case of stable language populations, mixtures of more than 1 language, as well as rapid language change. The 2 models also make distinct, empirically testable predictions about language change. Using historical data, we show that the SL model more faithfully replicates the dynamics of the evolution of Middle English.

Primate vocal behavior is often considered irrelevant in modeling human language evolution, mainly because of the caller's limited vocal control and apparent lack of intentional signaling. Here, we present the results of a long-term study on Campbell's monkeys, which has revealed an unrivaled degree of vocal complexity. Adult males produced six different loud call types, which they combined into various sequences in highly context-specific ways. We found stereotyped sequences that were strongly associated with cohesion and travel, falling trees, neighboring groups, nonpredatory animals, unspecific predatory threat, and specific predator classes. Within the responses to predators, we found that crowned eagles triggered four and leopards three different sequences, depending on how the caller learned about their presence. Callers followed a number of principles when concatenating sequences, such as nonrandom transition probabilities of call types, addition of specific calls into an existing sequence to form a different one, or recombination of two sequences to form a third one. We conclude that these primates have overcome some of the constraints of limited vocal control by combinatorial organization. As the different sequences were so tightly linked to specific external events, the Campbell's monkey call system may be the most complex example of 'proto-syntax' in animal communication known to date.

Abstract According to a controversial hypothesis, a characteristic unique to human language is recursion. Contradicting this hypothesis, it has been claimed that the starling, one of the two animal species tested for this ability to date, is able to distinguish acoustic stimuli ...

Symbolic gestures, such as pantomimes that signify actions (e.g., threading a needle) or emblems that facilitate social transactions (e.g., finger to lips indicating 'be quiet'), play an important role in human communication. They are autonomous, can fully take the place of words, and function as complete utterances in their own right. The relationship between these gestures and spoken language remains unclear. We used functional MRI to investigate whether these two forms of communication are processed by the same system in the human brain. Responses to symbolic gestures, to their spoken glosses (expressing the gestures' meaning in English), and to visually and acoustically matched control stimuli were compared in a randomized block design. General Linear Models (GLM) contrasts identified shared and unique activations and functional connectivity analyses delineated regional interactions associated with each condition. Results support a model in which bilateral modality-specific areas in superior and inferior temporal cortices extract salient features from vocal-auditory and gestural-visual stimuli respectively. However, both classes of stimuli activate a common, left-lateralized network of inferior frontal and posterior temporal regions in which symbolic gestures and spoken words may be mapped onto common, corresponding conceptual representations. We suggest that these anterior and posterior perisylvian areas, identified since the mid-19th century as the core of the brain's language system, are not in fact committed to language processing, but may function as a modality-independent semiotic system that plays a broader role in human communication, linking meaning with symbols whether these are words, gestures, images, sounds, or objects.

Studies of language change have begun to contribute to answering several pressing questions in cognitive sciences, including the origins of human language capacity, the social construction of cognition and the mechanisms underlying culture change in general. Here, we describe recent advances within a new emerging framework for the study of language change, one that models such change as an evolutionary process among competing linguistic variants. We argue that a crucial and unifying element of this framework is the use of probabilistic, data-driven models both to infer change and to compare competing claims about social and cognitive influences on language change.

The recent Bayesian approaches to language evolution and change seem to suggest that genetic biases can impact on the characteristics of language, but, at the same time, that its cultural transmission can partially free it from these same genetic constraints. One of the current debates centres on the striking differences between sampling and a posteriori maximising Bayesian learners, with the first converging on the prior bias while the latter allows a certain freedom to language evolution. The present paper shows that this difference disappears if populations more complex than a single teacher and a single learner are considered, with the resulting behaviours more similar to the sampler. This suggests that generalisations based on the language produced by Bayesian agents in such homogeneous single agent chains are not warranted. It is not clear which of the assumptions in such models are responsible, but these findings seem to support the rising concerns on the validity of the 'acquisitionist' assumption, whereby the locus of language change and evolution is taken to be the first language acquirers (children) as opposed to the competent language users (the adults).

Human languages form a distinct and largely independent class of cultural replicators with behaviour and fidelity that can rival that of genes. Parallels between biological and linguistic evolution mean that statistical methods inspired by phylogenetics and comparative biology are being increasingly applied to study language. Phylogenetic trees constructed from linguistic elements chart the history of human cultures, and comparative studies reveal surprising and general features of how languages evolve, including patterns in the rates of evolution of language elements and social factors that influence temporal trends of language evolution. For many comparative questions of anthropology and human behavioural ecology, historical processes estimated from linguistic phylogenies may be more relevant than those estimated from genes.

Throughout recorded history, sexually mature males have issued humorous insults in public. These averbal duelsa are thought to discharge aggressive dispositions, and to provide a way to compete for status and mating opportunities without risking physical altercations. But, is there evidence that such verbal duels, and sexual selection in general, played any role in the evolution of specific principles of language, syntax in particular? In this paper, concrete linguistic data and analysis will be presented which indeed point to that conclusion. The prospect will be examined that an intermediate form of aproto-syntaxa, involving aproto-Mergea, evolved in a context of ritual insult. This form, referred to as exocentric compound, can be seen as a aliving fossila of this stage of proto-syntax a providing evidence not only of ancient structure (syntax/semantics), but also arguably of sexual selection.

Word-order typology often uses the linear order of binary grammatical pairs in sentences to classify a language. The present paper proposes a method based on dependency treebanks as a typological means. This paper investigates 20 languages using treebanks with different sizes from 16 K to 1 million dependencies. The results show that some languages are more head-initial or head-final than others, but all contain head-initial and head-final elements. The 20 languages can be arranged on a continuum with complete head-initial and head-final patterns as the two ends. Some data about subjectaverb, objectaverb and adjectiveanoun are extracted from the treebanks for comparison with the typological studies based on the traditional means, the results are similar. The investigation demonstrates that the proposed method is valid for positioning a language in the typological continuum and the resources from computational linguistics can also be used in language typology.

In addition to its theoretical impact, the development of molecular biology has brought about the possibility of extraordinary historical progress in the study of phylogenetic classification of different species and human populations (especially cf. Cavalli Sforza et al., 1994, ...

According to Darwin [Darwin, CR. The descent of man, and selection in relation to sex. London: John Murray; 1871], the human musical faculty `must be ranked amongst the most mysterious with which he is endowed'. Music is a human cultural universal that serves no obvious adaptive purpose, making its evolution a puzzle for evolutionary biologists. This review examines Darwin's hypothesis of similarities between language and music indicating a shared evolutionary history. In particular, the fact that both are human universals, have phrase structure, and entail learning and cultural transmission, suggests that any theory of the evolution of language will have implications for the evolution of music, and vice versa. The argument starts by describing variable predispositional musical capabilities and the ontogeny of prosodic communication in human infants and young children, presenting comparative data regarding communication systems commonly present in living nonhuman primate species. Like language, the human music faculty is based on a suite of abilities, some of which are shared with other primates and some of which appear to be uniquely human. Each of these subcomponents may have a different evolutionary history, and should be discussed separately. After briefly considering possible functions of human music for language acquisition, the review ends by discussing the phylogenetic history of music. It concludes that many strands of evidence support Darwin's hypothesis of an intermediate stage of human evolutionary history, characterized by a communication system that resembled music more closely than language, but was identical to neither. This pre-linguistic system, which could probably referred to as `prosodic protolanguage', provided a precursor for both modern language and music.

Zipf's discovery that word frequency distributions obey a power law established parallels between biological and physical processes, and language, laying the groundwork for a complex systems perspective on human communication. More recent research has also identified scaling regularities in the dynamics underlying the successive occurrences of events, suggesting the possibility of similar findings for language as well.

By considering frequent words in USENET discussion groups and in disparate databases where the language has different levels of formality, here we show that the distributions of distances between successive occurrences of the same word display bursty deviations from a Poisson process and are well characterized by a stretched exponential (Weibull) scaling. The extent of this deviation depends strongly on semantic type - a measure of the logicality of each word - and less strongly on frequency. We develop a generative model of this behavior that fully determines the dynamics of word usage.

Recurrence patterns of words are well described by a stretched exponential distribution of recurrence times, an empirical scaling that cannot be anticipated from Zipf's law. Because the use of words provides a uniquely precise and powerful lens on human thought and activity, our findings also have implications for other overt manifestations of collective human dynamics.

The iterated classification game (ICG) combines the classification game with the iterated learning model (ILM) to create a more realistic model of the cultural transmission of language through generations. It includes both learning from parents and learning from peers. Further, it eliminates some of the chief criticisms of the ILM: that it does not study grounded languages, that it does not include peer learning, and that it builds in a bias for compositional languages. We show that, over the span of a few generations, a stable linguistic system emerges that can be acquired very quickly by each generation, is compositional, and helps the agents to solve the classification problem with which they are faced. The ICG also leads to a different interpretation of the language acquisition process. It suggests that the role of parents is to initialize the linguistic system of the child in such a way that subsequent interaction with peers results in rapid convergence to the correct language.

Social learning is a source of behaviour for many species, but few use it as extensively as they seemingly could. In this article, I attempt to clarify our understanding of why this might be. I discuss the potential computational properties of social learning, then examine the phenomenon in nature through creating a taxonomy of the representations that might underly it. This is achieved by first producing a simplified taxonomy of the established forms of social learning, then describing the primitive capacities necessary to support them, and finally considering which of these capacities we actually have evidence for. I then discuss theoretical limits on cultural evolution, which include having sufficient information transmitted to support robust representations capable of supporting variation for evolution, and the need for limiting the extent of social conformity to avoid ecological fragility. Finally, I show how these arguments can inform several key scientific questions, including the uniqueness of human culture, the long lifespans of cultural species, and the propensity of animals to seemingly have knowledge about a phenomenon well before they will act upon it.

In this paper, we propose two agent-based simulation models for the evolution of language in the framework of evolutionary language games. The theory of evolutionary language games arose from the union of evolutionary game theory, introduced by the English biologist John Maynard Smith, and language games, developed by the Austrian philosopher Ludwig Wittgenstein. The first model proposed is based on Martin Nowak's work and is designed to reproduce and verify (or refute) the results Nowak obtained in his simplest mathematical model. For the second model, we extend the previous one with the introduction of a world where the languages live and evolve, and which influences interactions among individuals. The main goal of this research is to present a model which shows how the presence of a topological structure influences the communication among individuals and contributes to the emergence of clusters of different languages.

The majority of the extant languages have one of three dominant basic word orders: SVO, SOV or VSO. Various hypotheses have been proposed to explain this word order bias, including the existence of a universal grammar, the learnability imposed by cognitive constraints, the descent of modern languages from an ancestral protolanguage, and the constraints from functional principles. We run simulations using a multi-agent computational model to study this bias. Following a local order approach, the model simulates individual language processing mechanisms in production and comprehension. The simulation results demonstrate that the semantic structures that a language encodes can constrain the global syntax, and that local syntax can help trigger bias towards the global order SOV/SVO (or VOS/OVS).

It is widely assumed that language in some form or other originated by piggybacking on pre-existing learning mechanism not dedicated to language. Using evolutionary connectionist simulations, we explore the implications of such assumptions by determining the effect of constraints derived from an earlier evolved mechanism for sequential learning on the interaction between biological and linguistic adaptation across generations of language learners. Artificial neural networks were initially allowed to evolve ``biologically'' to improve their sequential learning abilities, after which language was introduced into the population. We compared the relative contribution of biological and linguistic adaptation by allowing both networks and language to change over time. The simulation results support two main conclusions: First, over generations, a consistent head-ordering emerged due to linguistic adaptation. This is consistent with previous studies suggesting that some apparently arbitrary aspects of linguistic structure may arise from cognitive constraints on sequential learning. Second, when networks were selected to maintain a good level of performance on the sequential learning task, language learnability is significantly improved by linguistic adaptation but not by biological adaptation. Indeed, the pressure toward maintaining a high level of sequential learning performance prevented biological assimilation of linguistic-specific knowledge from occurring.

We show that dolphin whistle types tend to be used in specific behavioral contexts, which is consistent with the hypothesis that dolphin whistle have some sort of ameaninga. Besides, in some cases, it can be shown that the behavioral context in which a whistle tends to occur or not occur is shared by different individuals, which is consistent with the hypothesis that dolphins are communicating through whistles. Furthermore, we show that the number of behavioral contexts significantly associated with a certain whistle type tends to grow with the frequency of the whistle type, a pattern that is reminiscent of a law of word meanings stating, as a tendency, that the higher the frequency of a word, the higher its number of meanings. Our findings indicate that the presence of Zipf's law in dolphin whistle types cannot be explained with enough detail by a simplistic die rolling experiment.

This article investigates probability distributions of the dependency relation extracted from a Chinese dependency treebank. The author shows the frequency distributions of dependency type, of word class both as a dependent and a governor, of verb as a governor, and of noun as a dependent. The fitting results reveal that most of the investigated distributions are excellently fitted with a modified right-truncated Zipf-Alekseev distribution. In the analysis of exponential regressions, most of the determination coefficients R2 are very good, which is an alternative evidence that the investigated distributions are fitted well.

The sound inventories of the world's languages self-organize themselves giving rise to similar cross-linguistic patterns. In this work we attempt to capture this phenomenon of self-organization, which shapes the structure of the consonant inventories, through a complex network approach. For this purpose we define the occurrence and co-occurrence networks of consonants and systematically study some of their important topological properties. A crucial observation is that the occurrence as well as the co-occurrence of consonants across languages follow a power law distribution. This property is arguably a consequence of the principle of preferential attachment. In order to support this argument we propose a synthesis model which reproduces the degree distribution for the networks to a close approximation. We further observe that the co-occurrence network of consonants show a high degree of clustering and subsequently refine our synthesis model in order to incorporate this property. Finally, we discuss how preferential attachment manifests itself through the evolutionary nature of language.

Abstract Axelrod's model for culture dissemination offers a nontrivial answer to the question of why there is cultural diversity given that people's beliefs have a tendency to become more similar to each other's as they interact repeatedly. The answer depends on the two control ...

Trudgill (2004) proposed that the emergence of New Zealand English, and of isolated new dialects generally, is purely deterministic. It can be explained solely in terms of the frequency of occurrence of particular variants and the frequency of interactions between different speakers in the society. Trudgill's theory is closely related to usage-based models of language, in which frequency plays a role in the representation of linguistic knowledge and in language change. Trudgill's theory also corresponds to a neutral evolution model Of language change. We use a mathematical model based on Croft's usage-based evolutionary framework for language change (Baxter, Blythe, Croft, \& McKane, 2006), and investigate whether Trudgill's theory is a plausible model of the emergence of new dialects. The results of our modeling indicate that determinism cannot be a sufficient mechanism for the emergence of a new dialect. Our approach illustrates the utility of mathematical modeling of theories and of empirical data for the study of language change.

Constituent structure is considered to be the very foundation of linguistic competence and often considered to be innate, yet we show here that it is derivable from the domain-general processes of chunking and categorization. Using modern and diachronic corpus data, we show that the facts support a view of constituent structure as gradient (as would follow from its source in chunking and categorization) and subject to gradual changes over time. Usage factors (i.e., repetition) and semantic factors both influence chunking and categorization and, therefore, influence constituent structure. We take as our example the complex prepositions of English, for instance, on top of, in back of, and in spite of, whose internal constituent structure has been much debated. From observing strong (but not absolute) usage trends in the corpus data, we find that these complex preposition sequences display varying degrees of emerging constituency. We conclude that constituent reanalysis, like language change generally, proceeds gradually.

Language is a complex adaptive system: Speakers are agents who interact with each other, and their past and current interactions feed into speakers' future behavior in complex ways. In this article, we describe the social cognitive linguistic basis for this analysis of language and a mathematical model developed in collaboration between researchers in linguistics and statistical physics. The model has led us to posit two mechanisms of selectionaneutral interactor selection and weighted interactor selectionain addition to neutral evolution and replicator selection (fitness). We describe current results in modeling language change in terms of neutral interactor selection and weighted interactor selection.

All natural languages rely on sentence-level form-meaning associations (i.e., linking rules) to encode propositional content about who did what to whom. Although these associations are recognized as foundational in many different theoretical frameworks (Goldberg, 1995, 2006; Lidz, Gleitman, & Gleitman, 2003; Pinker, 1984, 1989) and areaat least in principlealearnable (Allen, 1997; Morris, Cottrell, & Elman, 2000), very little empirical work has been done to establish that human participants are able to acquire them from the input. In the present work, we provided adult participants with 3 min worth of exposure to a novel syntactic construction and then tested to see what was learned. Experiment 1 established that participants are able to accurately deploy newly acquired linking rules in a forced-choice comprehension task, and that constructional knowledge largely persists over a 1-week period. In Experiment 2, participants were exposed to the linking rules immanent in one of two novel constructions and were asked to describe novel events using their exposure construction. The data indicate that participants were successful in using their exposure construction's linking rules in production, and that performance was equally good regardless of the specifics of the target linking pattern. These results indicate that linking rules can be learned relatively easily by adults, which, in turn, suggests that children may also be capable of learning them directly from the input.

Most current approaches to linguistic structure suggest that language is recursive, that recursion is a fundamental property of grammar, and that independent performance constraints limit recursive abilities that would otherwise be infinite. This article presents a usage-based perspective on recursive sentence processing, in which recursion is construed as an acquired skill and in which limitations on the processing of recursive constructions stem from interactions between linguistic experience and intrinsic constraints on learning and processing. A connectionist model embodying this alternative theory is outlined, along with simulation results showing that the model is capable of constituent-like generalizations and that it can fit human data regarding the differential processing difficulty associated with center-embeddings in German and cross-dependencies in Dutch. Novel predictions are furthermore derived from the model and corroborated by the results of four behavioral experiments, suggesting that acquired recursive abilities are intrinsically bounded not only when processing complex recursive constructions, such as center-embedding and cross-dependency, but also during processing of the simpler, right- and left-recursive structures.

Language is a product of both biological and cultural evolution. Clues to the origins of key structural properties of language can be found in the process of cultural transmission between learners. Recent experiments have shown that iterated learning by human participants in the laboratory transforms an initially unstructured artificial language into one containing regularities that make the system more learnable and stable over time. Here, we explore the process of iterated learning in more detail by demonstrating exactly how one type of structureacompositionalityaemerges over the course of these experiments. We introduce a method to precisely quantify the increasing ability of a language to systematically encode associations between individual components of meanings and signals over time and we examine how the system as a whole evolves to avoid ambiguity in these associations and generate adaptive structure.

This article presents an analysis of interactions in the usage, structure, cognition, coadaptation of conversational partners, and emergence of linguistic constructions. It focuses on second language development of English verb-argument constructions (VACs: VL, verb locative; VOL, verb object locative; VOO, ditransitive) with particular reference to the following: (a) Construction learning as concept learning following the general cognitive and associative processes of the induction of categories from experience of exemplars in usage obtained through coadapted micro-discursive interaction with conversation partners; (b) the empirical analysis of usage by means of corpus linguistic descriptions of native and nonnative speech and of longitudinal emergence in the interlanguage of second language learners; (c) the effects of the frequency and Zipfian type/token frequency distribution of exemplars within the Verb and other islands of the construction archipelago (e.g., [Subj V Obj Oblpath/loc]), by their prototypicality, their generic coverage, and their contingency of form-meaning-use mapping, and (d) computational (emergent connectionist) models of these various factors as they play out in the emergence of constructions as generalized linguistic schema.

Language has a fundamentally social function. Processes of human interaction along with domain-general cognitive processes shape the structure and knowledge of language. Recent research in the cognitive sciences has demonstrated that patterns of use strongly affect how language is acquired, is used, and changes. These processes are not independent of one another but are facets of the same complex adaptive system (CAS). Language as a CAS involves the following key features: The system consists of multiple agents (the speakers in the speech community) interacting with one another. The system is adaptive; that is, speakers' behavior is based on their past interactions, and current and past interactions together feed forward into future behavior. A speaker's behavior is the consequence of competing factors ranging from perceptual constraints to social motivations. The structures of language emerge from interrelated patterns of experience, social interaction, and cognitive mechanisms. The CAS approach reveals commonalities in many areas of language research, including first and second language acquisition, historical linguistics, psycholinguistics, language evolution, and computational modeling.

The evolution of language and the evolution of the brain are tightly interlinked. Language evolution represents a special kind of adaptation, in part because language is a complex behavior (as opposed to a physical feature) but also because changes are adaptive only to the extent that they increase either one's understanding of others, or one's understanding to others. Evolutionary changes in the human brain that are thought to be relevant to language are reviewed. The extent to which these changes are a cause or consequence of language evolution is a good question, but it is argued that the process may best be viewed as a complex adaptive system, in which cultural learning interacts with biology iteratively over time to produce language.

We introduce a class of stochastic models for the dynamics of two linguistic variants that are competing to become the single, shared convention within an unstructured community of speakers. Different instances of the model are distinguished by the way agents handle variability in the language (i.e., multiple forms for the same meaning). The class of models includes as special cases two previously-studied models of language dynamics, the Naming Game, in which agents tend to standardise on variants they have encountered most frequently, and the Utterance Selection Model, in which agents tend to preserve variability by uniform sampling of a pool of utterances. We reduce the full complexities of the dynamics to a single-coordinate stochastic model which allows the probability and time taken for speakers to reach consensus on a single variant to be calculated for large communities. This analysis suggests that in the broad class of models considered, consensus is formed in one of three generic ways, according to whether agents tend to eliminate, accentuate or sample neutrally the variability in the language. These different regimes are observed in simulations of the full dynamics, and for which the simplified model in some cases makes good quantitative predictions. We use these results, along with comparisons with related models, to conjecture the likely behaviour of more general models, and further make use of empirical data to argue that in reality, biases away from neutral sampling behaviour are likely to be small.

Since language is tied to cognition, we expect the linguistic structures to reflect patterns we encounter in nature and analyzed by physics. Within this realm we investigate the process of protolanguage acquisition, using analytical and tractable methods developed within physics. A protolanguage is a mapping between sounds and objects (or concepts) of the perceived world. This mapping is represented by a matrix and the linguistic interaction among individuals is described by a random matrix model. There are two essential parameters in our approach. The strength of the linguistic interaction $\beta$, which following Chomsky's tradition, we consider as a genetically determined ability, and the number $N$ of employed sounds (the lexicon size). Our model of linguistic interaction is analytically studied using methods of statistical physics and simulated by Monte Carlo techniques. The analysis reveals an intricate relationship between the innate propensity for language acquisition $\beta$ and the lexicon size $N$, $N \sim \exp(\beta)$. Thus a small increase of the genetically determined $\beta$ may lead to an incredible lexical explosion. Our approximate scheme offers an explanation for the biological affinity of different species and their simultaneous linguistic disparity.

We introduce a simple open-ended model that describes the emergence of a shared vocabulary. The ordering transition toward consensus is generated only by an agreement mechanism. This interaction defines a finite and small number of states, despite each individual having the ability to invent an unlimited number of new words. The existence of a phase transition is studied by analyzing the convergence times, the cognitive efforts of the agents and the scaling behavior in memory and time.

Statistical physics has proven to be a fruitful framework to describe phenomena outside the realm of traditional physics. Recent years have witnessed an attempt by physicists to study collective phenomena emerging from the interactions of individuals as elementary units in social structures. A wide list of topics are reviewed ranging from opinion and cultural and language dynamics to crowd behavior, hierarchy formation, human dynamics, and social spreading. The connections between these problems and other, more traditional, topics of statistical physics are highlighted. Comparison of model results with empirical data from social systems are also emphasized.

We consider two social consensus models, the AB-model and the Naming Game restricted to two conventions, which describe a population of interacting agents that can be in either of two equivalent states (A or B) or in a third mixed (AB) state. Proposed in the context of language competition and emergence, the AB state was associated with bilingualism and synonymy respectively. We show that the two models are equivalent in the mean field approximation, though the differences at the microscopic level have non-trivial consequences. To point them out, we investigate an extension of these dynamics in which confidence/trust is considered, focusing on the case of an underlying fully connected graph, and we show that the consensus-polarization phase transition taking place in the Naming Game is not observed in the AB model. We then consider the interface motion in regular lattices. Qualitatively, both models show the same behavior: a diffusive interface motion in a one-dimensional lattice, and a curvature driven dynamics with diffusing stripe-like metastable states in a two-dimensional one. However, in comparison to the Naming Game, the AB-model dynamics is shown to slow down the diffusion of such configurations.

Abstract: A key challenge for theories of language evolution is to explain why language is the way it is and how it came to be that way. It is clear that how we learn and use language is governed by genetic constraints. However, the nature of these innate constraints has been ...

We show that the law of brevity, i.e. the tendency of words to shorten as their frequency increases, is also found in dolphin surface behavioral patterns. As far as we know, this is the first evidence of the law in another species, suggesting that coding efficiency is not unique to humans.

It has been argued that the actual distribution of word frequencies could be reproduced or explained by generating a random sequence of letters and spaces according to the so-called intermittent silence process. The same kind of process could reproduce or explain the counts of other kinds of units from a wide range of disciplines. Taking the linguistic metaphor, we focus on the frequency spectrum, i.e., the number of words with a certain frequency, and the vocabulary size, i.e., the number of different words of text generated by an intermittent silence process. We derive and explain how to calculate accurately and efficiently the expected frequency spectrum and the expected vocabulary size as a function of the text size.

In neural network research on language, the existence of discrete combinatorial rule representations is commonly denied. Combinatorial capacity of networks and brains is rather attributed to probability mapping and pattern overlay. Here, we demonstrate that networks incorporating relevant features of neuroanatomical connectivity and neuronal function give rise to discrete neuronal circuits that store combinatorial information and exhibit a function similar to elementary rules of grammar. Key properties of these networks are rich auto- and hetero-associative connectivity, availability of sequence detectors similar to those found in a range of animals, and unsupervised Hebbian learning. Input of specific word sequences establishes sequence detectors in the network, and substitutions of words and larger string segments from one syntactic category, occurring in the context of elements of a second syntactic class, lead to binding between them into neuronal assemblies. Critically, these newly formed aggregates of sequence detectors now respond in a discrete generalizing fashion when members of specific substitution classes of string elements are combined with each other. The discrete combinatorial neuronal assemblies (DCNAs) even respond in the same way to learned strings and to word sequences that never appeared in the input but conform to a rule. We also show how combinatorial information interacts with information about functional and anatomical properties of the brain in the emergence of discrete neuronal circuits that may implement rules and discuss the model in the wider context of brain mechanism for syntax and grammar. Implications for the evolution of human language are discussed in closing.

In the last few years, researchers have begun to investigate the emergence of novel forms of human communication in the laboratory. I survey this growing line of research, which may be called experimental semiotics, from three distinct angles. First, I situate the new approach in its theoretical and historical context. Second, I review a sample of studies that exemplify experimental semiotics. Third, I present an empirical study that illustrates how the new approach can help us understand the socio-cognitive underpinnings of human communication. The main conclusion of the paper will be that, by reproducing micro samples of historical processes in the laboratory, experimental semiotics offers new powerful tools for investigating human communication as a form of joint action.

Two decades of attempts to model the emergence of language as a collective cognitive activity have demonstrated a number of principles that might have been part of the historical process that led to language. Several models have demonstrated the emergence of structure in a symbolic medium, but none has demonstrated the emergence of the capacity for symbolic representation. The current shift in cognitive science toward theoretical frameworks based on embodiment is already furnishing computational models with additional mechanisms relevant to the emergence of symbolic language. An analysis of embodied interaction among captive, but not human-enculturated, bonobo chimpanzees reveals a number of additional features of embodiment that are relevant to the emergence of symbolic language, but that have not yet been explored in computational simulation models; for example, complementarity of action in addition to imitation, iconic in addition to indexical gesture, coordination among multiple sensory and perceptual modalities, and the orchestration of intra- and inter-individual motor coordination. The bonobos provide an evolutionarily plausible intermediate stage in the development of symbolic expression that can inform efforts to model the emergence of symbolic language.

Phylogenetic methods have recently been applied to studies of cultural evolution. However, it has been claimed that the large amount of horizontal transmission that sometimes occurs between cultural groups invalidates the use of these methods. Here, we use a natural model of linguistic evolution to simulate borrowing between languages. The results show that tree topologies constructed with Bayesian phylogenetic methods are robust to realistic levels of borrowing. Inferences about divergence dates are slightly less robust and show a tendency to underestimate dates. Our results demonstrate that realistic levels of reticulation between cultures do not invalidate a phylogenetic approach to cultural and linguistic evolution.

The evolution of languages provides a unique opportunity to study human population history. The origin of Semitic and the nature of dispersals by Semitic-speaking populations are of great importance to our understanding of the ancient history of the Middle East and Horn of Africa. Semitic populations are associated with the oldest written languages and urban civilizations in the region, which gave rise to some of the world's first major religious and literary traditions. In this study, we employ Bayesian computational phylogenetic techniques recently developed in evolutionary biology to analyse Semitic lexical data by modelling language evolution and explicitly testing alternative hypotheses of Semitic history. We implement a relaxed linguistic clock to date language divergences and use epigraphic evidence for the sampling dates of extinct Semitic languages to calibrate the rate of language evolution. Our statistical tests of alternative Semitic histories support an initial divergence of Akkadian fromancestral Semitic over competing hypotheses (e.g. an African origin of Semitic). We estimate an Early Bronze Age origin for Semitic approximately 5750 years ago in the Levant, and further propose that contemporary Ethiosemitic languages of Africa reflect a single introduction of early Ethiosemitic from southern Arabia approximately 2800 years ago.

The question as to whether cultures evolve in a manner analogous to that of genetic evolution can be addressed by attempting to reconstruct population histories using cultural data. As others have argued, this can only succeed if cultures are isolated enough to maintain and pass on a central core of traditions that can be modified over time. In this study we used a set of cultural data (canoe design traits from Polynesia) to look for the kinds of patterns and relationships normally found in population genetic studies. After developing new techniques to accommodate the peculiarities of cultural data, we were able to infer an ancestral region (Fiji) and a sequence of cultural origins for these Polynesian societies. In addition, we found evidence of cultural exchange, migration and a serial founder effect. Results were stronger when analyses were based on functional traits (presumably subject to natural selection and convergence) rather than symbolic or stylistic traits (probably subject to cultural selection for rapid divergence). These patterns strongly suggest that cultural evolution, while clearly affected by cultural exchange, is also subject to some of the same processes and constraints as genetic evolution.

We invoke the Tsallis entropy formalism, a nonextensive entropy measure, to include some degree of non-locality in a neural network that is used for simulation of novel word learning in adults. A generalization of the gradient descent dynamics, realized via nonextensive cost functions, is used as a learning rule in a simple perceptron. The model is first investigated for general properties, and then tested against the empirical data, gathered from simple memorization experiments involving two populations of linguistically different subjects. Numerical solutions of the model equations corresponded to the measured performance states of human learners. In particular, we found that the memorization tasks were executed with rather small but population-specific amounts of nonextensivity, quantified by the entropic index q. Our findings raise the possibility of using entropic nonextensivity as a means of characterizing the degree of complexity of learning in both natural and artificial systems.

The naming game model characterizes the main evolutionary features Of languages or more generally of communication systems. Very recently, the combination of complex networks and the naming game has received Much attention and the influences of various topological properties on the corresponding dynamical behavior have been widely studied. In this paper, we investigate the naming game on small-world geographical networks. The small-world geographical networks are constructed by randomly adding links to two-dimensional regular lattices, and it is found that the convergence time is a nonmonotonic function of the geographical distance of randomly added shortcuts. This phenomenon indicates that. although a long geographical distance of the added shortcuts favors consensus achievement, too long a geographical distance of the added shortcuts inhibits the convergence process, making it even slower than the moderates.

Human history leaves fingerprints in human languages. Little is known over language evolution and its study is of great importance. Here, we construct a simple stochastic model and compare its results to statistical data of real languages. The model is based on the recent findings that language changes occur independently of the population size. We find agreement with the data additionally assuming that languages may be distinguished by having at least one among a finite, small number of different features. This finite set is used also in order to define the distance between two languages, similarly to linguistics tradition since Swadesh.

Traditionally, grammaticalisation has been described as being based on phenomena specific to language such as metaphorical extension or reanalysis. This characterisation is somewhat in contrast to claims that grammaticalisation is involved in the much more general process of the initial emergence of language. In this article, we provide a unified analysis of both the metaphor-based and the reanalysis-based account of grammaticalisation which is grounded in the cognitive mechanisms underlying ostensive-inferential communication. We are thus able to show that the process of grammaticalisation is an instantiation of a domain-general pre-linguistic phenomenon.

The evolution of color categorization is investigated using artificial agent population categorization games, by modeling observer types using Farnsworth-Munsell 100 Hue Test performance to capture human processing constraints on color categorization. Homogeneous populations of both normal and dichromat agents are separately examined. Both types of populations produce near-optimal categorization solutions. While normal observers produce categorization solutions that show rotational invariance, dichromats' solutions show symmetry-breaking features. In particular, it is found that dichromats' local confusion regions tend to repel color category boundaries and that global confusion pairs attract category boundaries. The trade-off between these two mechanisms gives rise to population categorization solutions where color boundaries are anchored to a subset of locations in the stimulus space. A companion paper extends these studies to more realistic, heterogeneous agent populations [J. Opt. Soc. Am. A 26, 1424-1436 (2009)].

The evolution of color categorization is investigated using computer simulations of agent population categorization games. Various realistic observer types are implemented based on Farnsworth-Munsell 100 Hue Test human performance data from normal and anomalous trichromats, dichromats, and humans with four retinal photopigments. Results show that (i) a small percentage of realistically modeled deficient agents greatly affects the shared categorization solutions of the entire population in terms of color category boundary locations; (ii) for realistically modeled populations, dichromats have the strongest influence on the color categorization; their characteristic forms of color confusion affect (i.e., attract or repel) color boundary locations and accord with our findings for homogeneous dichromat populations [J. Opt. Soc. Am. A26, 1414-1423 (2009)]; (iii) adding anomalous trichromats or trichromat aexpertsa does not destabilize the solutions or substantially alter solution structure. The results provide insights regarding the mechanisms that may constrain universal tendencies in human color categorization systems.

Attempts to describe language competition and extinction in a mathematical way have enjoyed increased popularity recently. In this paper I review recent modeling approaches and, based on these findings, propose a model of reaction-diffusion type. I analyze the dynamics of interactions of a population with two monolingual groups and a group that is bilingual in these two languages. The results show that demographic factors, such as population growth or population dispersal, play an important role in the competition dynamic. Furthermore, I consider the impact of two strategies for language maintenance: adjusting the status of the endangered language and adjusting the availability of monolingual and bilingual educational resources.

In this article I provide a review of studies that have modeled interactions between language evolution and demographic processes. The models are classified in terms of three different approaches: analytical modeling, agent-based analytical modeling, and agent-based cognitive modeling. I show that these approaches differ in the complexity of interactions that they can handle and that the agent-based cognitive models allow for the most detailed and realistic simulations. Thus readers are provided with a guideline for selecting which approach to use for a given problem. The analytical models are useful for studying interactions between demography and language evolution in terms of high-level processes; the agent-based analytical models are good for studying such interactions in terms of social dynamics without bothering too much about the cognitive mechanisms of language processing; and the agent-based cognitive models are best suited for the study of the interactions between the complex sociocognitive mechanisms underlying language evolution.

Previous empirical studies of population size and language change have produced equivocal results. We therefore address the question with a new set of lexical data from nearly one half of the world's languages. We first show that relative population sizes of modern languages can be extrapolated to ancestral languages, albeit with diminishing accuracy, up to several thousand years into the past. We then test for an effect of population against the null hypothesis that the ultrametric inequality is satisfied by lexical distances among triples of related languages. The test shows mainly negligible effects of population, the exception being an apparently faster rate of change in the larger of two very closely related variants. A possible explanation for the exception may be the influence on emerging standard (or cross-regional) variants from speakers that shift from different dialects to the standard. Our results strongly indicate that the sizes of speaker populations do not in and of themselves determine rates of language change. Comparison of this empirical finding with previously published computer simulations suggests that the most plausible model for language change is one in which changes propagate at a local level in a type of network where the individuals have different degrees of connectivity.

We examine a naming game with two agents trying to establish a common vocabulary for n objects. Such efforts lead to the emergence of language that allows for an efficient communication and exhibits some degree of homonymy and synonymy. Although homonymy reduces the communication efficiency, it seems to be a dynamical trap that persists for a long, and perhaps indefinite, time. On the other hand, synonymy does not reduce the efficiency of communication, but appears to be only a transient feature of the language. Thus, in our model the role of synonymy decreases and in the long-time limit it becomes negligible. A similar rareness of synonymy is observed in present natural languages. The role of noise, that distorts the communicated words, is also examined. Although, in general, the noise reduces the communication efficiency, it also regroups the words so that they are more evenly distributed within the available ``verbal'' space.

Almost all language networks in word and syntactic levels are small-world and scale-free. This raises the questions of whether a language network in deeper semantic or cognitive level also has the similar properties. To answer the question, we built up a Chinese semantic network based on a treebank with semantic role (argument structure) annotation and investigated its global statistical properties. The results show that although semantic network is also small-world and scale-free, it is different from syntactic network in hierarchical structure and K-Nearest-Neighbor correlation.

This article describes a method for calculating the adependency distancea between the words in a text a i.e. the number of words that separate each word from the word on which it depends syntactically a and reports the results of applying this method to a Chinese treebank. This study shows that Chinese dependencies tend strongly to be governor-final and that the mean dependency distance of words is much higher for Chinese than for other languages that have been studied including English, German and Japanese. It is unclear whether this difference means that Chinese is syntactically more difficult to process.

Since language is a biological trait, it is necessary to investigate its evolution, development, and functions, along with the mechanisms that have been set aside, and are now recruited, for its acquisition and use. It is argued here that progress toward each of these goals can ...

The standard view of classical cognitive science stated that cognition consists in the manipulation of language-like structures according to formal rules. Since cognition is linguistic in itself, according to this view language is just a complex communication system and does not influence cognitive processes in any substantial way. This view has been criticized from several perspectives and a new framework (Embodied Cognition) has emerged that considers cognitive processes as non-symbolic and heavily dependent on the dynamical interactions between the cognitive system and its environment. But notwithstanding the successes of the embodied cognitive science in explaining low-level cognitive behaviors, it is still not clear whether and how it can scale up for explaining high-level cognition. In this paper we argue that this can be done by considering the role of language as a cognitive tool: i.e. how language transforms basic cognitive functions in the high-level functions that are characteristic of human cognition. In order to do that, we review some computational models that substantiate this view with respect to categorization and memory. Since these models are based on a very rudimentary form of non-syntactic language we argue that the use of language as a cognitive tool might have been an early discovery in hominid evolution, and might have played a substantial role in the evolution of language itself.

Language development in children provides a window to understand the transition from protolanguage to language. Here we present the first analysis of the emergence of syntax in terms of complex networks. A previously unreported, sharp transition is shown to occur around two years of age from a (pre-syntactic) tree-like structure to a scale-free, small world syntax network. The development of these networks thus reveals a nonlinear dynamical pattern where the global topology of syntax graphs shifts from a hierarchical, tree-like pattern, to a scale-free organization. Such change seems difficult to be explained under a self-organization framework. Instead, it actually supports the presence of some underlying innate component, as early suggested by some authors.

A unique hallmark of human language is that it uses signals that are both learnt and symbolic. The emergence of such signals was therefore a defining event in human cognitive evolution, yet very little is known about how such a process occurs. Previous work provides some insights on how meaning can become attached to form, but a more foundational issue is presently unaddressed. How does a signal signal its own signalhood? That is, how do humans even know that communicative behaviour is indeed communicative in nature? We introduce an experimental game that has been designed to tackle this problem. We find that it is commonly resolved with a bootstrapping process, and that this process influences the final form of the communication system. Furthermore, sufficient common ground is observed to be integral to the recognition of signalhood, and the emergence of dialogue is observed to be the key step in the development of a system that can be employed to achieve shared goals.

The regularization of linguistic structures by learners has played a key role in arguments for strong innate constraints on language acquisition, and has important implications for language evolution. However, relating the inductive biases of learners to regularization behavior in laboratory tasks can be challenging without a formal model. In this paper we explore how regular linguistic structures can emerge from language evolution by iterated learning, in which one person's linguistic output is used to generate the linguistic input provided to the next person. We use a model of iterated learning with Bayesian agents to show that this process can result in regularization when learners have the appropriate inductive biases. We then present three experiments demonstrating that simulating the process of language evolution in the laboratory can reveal biases towards regularization that might not otherwise be obvious, allowing weak biases to have strong effects. The results of these experiments suggest that people tend to regularize inconsistent word-meaning mappings, and that even a weak bias towards regularization can allow regular languages to be produced via language evolution by iterated learning.

The regularization of linguistic structures by learners has played a key role in arguments for strong innate constraints on language acquisition, and has important implications for language evolution. However, relating the inductive biases of learners to regularization behavior in laboratory tasks can be challenging without a formal model. In this paper we explore how regular linguistic structures can emerge from language evolution by iterated learning, in which one person's linguistic output is used to generate the linguistic input provided to the next person. We use a model of iterated learning with Bayesian agents to show that this process can result in regularization when learners have the appropriate inductive biases. We then present three experiments demonstrating that simulating the process of language evolution in the laboratory can reveal biases towards regularization that might not otherwise be obvious, allowing weak biases to have strong effects. The results of these experiments suggest that people tend to regularize inconsistent word-meaning mappings, and that even a weak bias towards regularization can allow regular languages to be produced via language evolution by iterated learning.

The region of the ancient Sahul continent (present day Australia and New Guinea, and surrounding islands) is home to extreme linguistic diversity. Even apart from the huge Austronesian language family, which spread into the area after the breakup of the Sahul continent in the Holocene, there are hundreds of languages from many apparently unrelated families. On each of the subcontinents, the generally accepted classification recognizes one large, widespread family and a number of unrelatable smaller families. If these language families are related to each other, it is at a depth which is inaccessible to standard linguistic methods. We have inferred the history of structural characteristics of these languages under an admixture model, using a Bayesian algorithm originally developed to discover populations on the basis of recombining genetic markers. This analysis identifies 10 ancestral language populations, some of which can be identified with clearly defined phylogenetic groups. The results also show traces of early dispersals, including hints at ancient connections between Australian languages and some Papuan groups (long hypothesized, never before demonstrated). Systematic language contact effects between members of big phylogenetic groups are also detected, which can in some cases be identified with a diffusional or substrate signal. Most interestingly, however, there remains striking evidence of a phylogenetic signal, with many languages showing negligible amounts of admixture.

Grammars... refer to real structures, though not to psychologically real structures in the processing sense... a grammar is a description of our knowledge of a social institution-the language-and because of this basis in social or institutional reality, rather than in cognitive ...

In this review and position paper we explore the neural substrates for manual specialization and their possible connection with language and speech. We focus on two contrasting hypotheses of the origins of language and manual specialization: the language-first scenario and the tool-use-first scenario. Each one makes specific predictions about hand-use in non-human primates, as well as about the necessity of an association between speech adaptations and population-level right-handedness in the archaeological and fossil records. The concept of handedness is reformulated for archaeologists in terms of manual role specialization, using Guiard's model of asymmetric bimanual coordination. This focuses our attention on skilled bimanual tasks in which both upper limbs play complementary roles. We review work eliciting non-human primate hand preferences in co-ordinated bimanual tasks, and relevant archaeological data for estimating the presence or absence of a population-level bias to the right hand as the manipulator in extinct hominin species and in the early prehistory of our own species.

A fundamental, universal property of human language is that its phonology is combinatorial. That is, one can identify a set of basic, distinct units (phonemes, syllables) that can be productively combined in many different ways. In this paper, we develop a methodological framework based on evolutionary game theory for studying the evolutionary transition from holistic to combinatorial signal systems, and use it to evaluate a number of existing models and theories. We find that in all problematic linguistic assumptions are made or crucial components of evolutionary explanations are omitted. We present a novel model to investigate the hypothesis that combinatorial phonology results from optimizing signal systems for perceptual distinctiveness. Our model differs from previous models in three important respects. First, signals are modeled as trajectories through acoustic space; hence, both holistic and combinatorial signals have a temporal structure. Second, acoustic distinctiveness is defined in terms of the probability of confusion. Third, we show a path of ever increasing fitness from unstructured, holistic signals to structured signals that can be analyzed as combinatorial. On this path, every innovation represents an advantage even if no-one else in a population has yet obtained it.

Starting from the assumption that syntactic categories can be universally identified or correlated (whether or not they are instantiated in every languages), this chapter investigates the relation between them and their semantic interpretations, focusing on the nominal domain. It considers two possibilities: first, there might be a universal stock of categories from which individual languages might draw; second, there may be hypotheses that all languages must instantiate particular categories. The situation in phonological systems is illuminating: the stock of possible sounds is given by a general theory of phonetics-phonology, but not all of the categories need be utilized in every language.

This chapter argues that linguistic universals need to be understood in terms of a model of language that incorporates both learned statistical patterns (ahabitsa) and derivations (arulesa). It presents an Analysis by Synthesis model, where sentences are initially given a basic semantic interpretation based on canonical statistical patterns of syntax, but sentences are also at the same time assigned a separate derivation, reflecting the syntactic relationship between constituents. It proposes a universal constraint on language that is necessary for the model to link statistical patterns with syntactic derivations. This constraintathe acanonical form constraintaarequires that all languages must have a set of statistically dominant structural patterns indicating the mapping between syntactic constructions and their meanings. Moreover, it should be possible to approximate the meaning of complex derivations in terms of such canonical patterns without recourse to a full derivational analysis. This approach as complementary to the Minimalist program in that it seeks to determine what is minimally required to explain language acquisition and use.

This chapter discusses the usage-based theory of language. It argues that from this perspective there are very few synchronic universals in the sense of features that can be found in all languages. The only synchronic universal is that all languages have at least some minimal derivational morphology. It further argues that language change has to be taken into account in order to understand language universals. Factors relating to language useasuch as frequency of usagealead to grammaticalization, which tends to follow specific developmental paths. It suggests that language universals may be best viewed in terms of such unidirectional paths of linguistic change, driven by constraints arising from domain-general processes rather than ones that are specific to language.

This chapter begins with a brief discussion of the general perspective in the linguistic community on language universals. It then presents an overview of the subsequent chapters in this book. This is followed by a discussion of the importance of interdisciplinary research and a multidisciplinary approach towards understanding language universals.

Using the preceding chapter as a point of departure, this chapter offers a critical perspective on the notion of innate universals. It presents a 'minimal nativism' view, according to which a brain area should be seen as embodying a kind of language universal if it is genetically predisposed toward fulfilling a certain sufficiently general linguistic function, for example by virtue of its strategic connectivity. On this view, Broca's area could still count as the brain locus of a linguistic universal, even if it supports other functions beside language.

This chapter argues that the existence of universals in language would only be surprising if the rest of cognition, as well as the world at large, were unstructured. Given that the world is in some sense and to some extent predictable, universals should be sought in the structure of information it presents to the language system. A productive approach to the study of language universals could follow the lead of biology, where looking at the interplay of evolution and development is proving particularly effective.

This chapter examines synchronic cross-linguistic patterns in grammars and language use. It proposes that avariation-defininga universals delimit the scope of possible variation across languages. Examples of such universals include the Greenbergian implicational universals and the parameters in the Government-Binding tradition. It argues that variation-defining universals should be understood in terms of performance principles. It further suggests these same performance principles govern variation of structures within languages, dictating that following a verb, short prepositional phrases should precede long prepositional phrases.

This chapter focuses on linguistic universals embodied in Universal Grammar (UG), a characterization of the innate properties of the language faculty. Approaching language universals from a minimalist perspective, it begins by contrasting I-universals (innate properties of UG) with E-universals (universals in the Greenbergian tradition). It argues that even if every language displayed some property P, it would not imply that P is an I-universal, whereas P would be considered an E-universal. The chapter considers the relative importance of the following three factors in accounting for I-universals: (a) genetic endowment, (b) experience, and (c) language-independent principles. It concludes that the minimalist perspective suggests that I-universalsathe key properties of UGamay not be genetically encoded but instead may derive from language-independent principles of good design.

This chapter takes the goal of linguistics to be an explanation of how whole languages get to be the way they are. It shows that some insight can be gained into the forces shaping languages by considering them as products of a historical spiral involving both acquisition and production, learning and speaking, and occasionally innovating, over many generations.

Hauser, Chomsky, and Fitch (HCF) proposed that recursion is the only thing that distinguishes language (a) from other human capacities, and (b) from the capacities of animals. These factors are independent. The narrow faculty of language might include more than recursion, falsifying (a). Or it might consist only of recursion, although parts of the broad faculty might be uniquely human as well, falsifying (b). This chapter presents a view that is contrasted with HCF's above. It shows that there is considerably more of language that is special, though still a plausible product of the processes of evolution. It assesses the key bodies of evidence, coming to a different reading from HCF's. The chapter organizes the discussion by distinguishing the conceptual, sensorimotor, and specifically linguistic aspects of the broad language faculty in turn.

This chapter asks how the kind of language universals discussed in the previous chapters might be instantiated in human brains. It distinguishes between ashallowa and adeepa universals in cognition, the former being due to abstract computational properties, and the latter to properties of the neural architecture that supports the function in question, such as language. It argues that shallow universals that are a matter of consensus in the linguistic community should be studied from a neurodevelopmental standpoint to seek their deep (i.e., biologically meaningful) counterparts. Based on an extensive survey of genetic, anatomical, and imaging data, the chapter suggests that the specific architecture of local brain areas (such as Broca's area) is not genetically predetermined but instead emerges as a result of its role and activity, given its particular location in functional networks. A neurodevelopmental account of putative language universals is most likely to be based on organization and interaction of nonlinguistic 'ingredient processes'.

Natural languages share common features known as linguistic universals but the nature and origin of these features remain controversial. Generative approaches propose that linguistic universals are defined by a set of innately specified linguistic constraints in universal grammar (UG). The UG hypothesis is primarily supported by Poverty of Stimulus (POS) arguments that posit that the structure of language cannot be learned from exposure to the linguistic environment. This chapter reviews recent computational and empirical research in statistical learning that raises serious questions about the basic assumptions of POS arguments. More generally, these results question the validity of UG as the basis for linguistic universals. As an alternative, the chapter proposes that linguistic universals should be viewed as functional features of language, emerging from constraints on statistical learning mechanisms themselves and from general functional and pragmatic properties of communicative interactions.

This chapter reports on research showing that it may be a universal structural property of human languages that they fall into a class of languages defined by mildly context-sensitive grammars. It also investigates the issue of whether there are properties of language that are needed to guarantee that it is learnable. It suggests that languages are learnable if they have a finite Vapnik-Chervonenkis (VC) dimension (where the VC dimension provides a combinatorial measure of complexity for a set of languages). Informally, a finite VC dimension requires that there be restrictions on the set of languages to be learned such that they do not differ from one another in arbitrary ways. These restrictions can be construed as universals that are required for language to be learnable (given formal language learnability theory). The chapter concludes by pointing out that formalizations of the semantic contribution (e.g., compositionality) to language learning might yield further insight into language universals.

This chapter attempts to link the specific form taken by the universal grammatical mechanism that projects the finite lexicon of any given language onto the infinite set of strings of words paired with meanings that constitute that language to a more primitive capacity for planning, or constructing sequences of actions that culminate in an intended goal. A central question in defining this system is that of how action representations can be learned from interaction with the physical world. The formation of novel plans from such elementary actions requires two fundamental operations of composition, or sequencing, and type-raising, or mapping objects in a situation into their affordances, or contextually supported actions. It is argued that operations related to composition and type-raising determine the universal grammatical mechanism that projects language-specific lexicons onto the sentences of the language. This observation suggests that the language faculty is in evolutionary and developmental terms attached to a more primitive planning mechanism to which it is formally entirely transparent.

In order to make distinctive speech sounds, it is necessary to control two separate acoustic cavities. There has been a longstanding debate about whether a lowered larynx is essential for this. Lieberman and Crelin have used it as an argument against speech in Neanderthals. This claim is controversial, not only for paleontological reasons, but also because researchers do not agree on the need of a lowered larynx for distinctive speech. Researchers using similar methods (computer modeling) arrive at opposite conclusions. The problem is that one needs to take articulatory and anatomical constraints into account when investigating the acoustic implications of vocal tract morphology.

In order to study what the effect of lowering the larynx is, a reimplementation of Mermelstein s vocal tract model has been made. This is a computer model of the geometry of the (human male) vocal tract, whose controls correspond to the actions of the muscles involved in speech. This model was used to explore the possible articulations and the corresponding acoustic signals of different vocal tract geometries.

Experiments were run with the original male model, a model of the female vocal tract and a model that is a combination of these two tracts. It was also found that the female vocal tract is better than the male one. This observation was confirmed by a reanalysis of the data from the Peterson and Barney study. This establishes an evolutionary advantage of a vocal tract that has a pharyngeal and oral cavity of equal length (as in the female tract). It has a larger signaling space than the male tract. Males probably had evolutionary advantage from size exaggeration, as proposed by Fitch. It must be noted however, that the differences found so far are significant but small.

This survey is meant to explore the structure and dynamics of natural languages in the framework of complex networks. We begin with a description of lexical networks, where the nodes are words and edges represent lexical relationship between two words such as phonetic and semantic similarity. This is followed by an overview of various networks where again the nodes are the words, but unlike the case of lexical networks, the edges represent their co-occurrences in similar context. These networks are representations of the interactions among words as governed by the grammar rules of a language. Next we discuss the properties of phonological networks, where the nodes are sub-lexical units such as phonemes or syllables. Applications of linguistic networks in Natural Language Processing (NLP) and Information Retrieval (IR) are also discussed. We conclude the survey by enumerating some open problems in the area of linguistic networks.

This article discusses human language in the context of the major evolutionary transitions in the history of life. Because of its unique structure, language enables the transmission of unlimited cultural information in our species. Understanding its evolution is therefore an important topic for both cognitive science and evolutionary theory more widely. This article highlights points of consensus on what is crucial for progress in this area: understanding preadaptations; the necessity for interdisciplinarity; and the importance of modeling, comparative approaches, and genetics. It also discusses current controversies: biological versus cultural evolution, vocal versus manual origins, and the nature of protolanguage.

We use agent-based Monte Carlo simulations to address the problem of language choice dynamics in a tripartite community which is linguistically homogeneous but politically divided. We observe the process of non-local pattern formation that causes populations to self-organize into stable antagonistic groups due to the local dynamics of attraction and influence between individual computational agents. Our findings uncover some of the unique properties of opinion formation in social groups when the process is affected by asymmetric noise distribution, unstable inter-group boundaries, and different migratory behaviors. Although we focus on one particular study, the proposed stochastic dynamic models can be easily generalized and applied to investigate the evolution of other complex and nonlinear features of human collective behavior.

Computational modelling has proven a useful method to study the emergence of language-like communication systems. However, most existing models abstract away from the two facts that (1) lan- guage use exhibits pragmatic plasticity1 and (2) linguistic knowledge is an integral part of human conceptual knowledge. This paper introduces a basic architecture for a model that overcomes these shortcomings by incorporating elements of Relevance Theory and Cognitive Semantics.

For a long time, human language has been assumed to be genetically determined and therefore the product of biological evolution. It is only within the last decade that researchers have begun to investigate more closely the domain-general cognitive mechanisms of cultural evolution as an alternative explanation for the origins of language. Most of this more recent work focuses on the role of imperfect cultural transmission and abstracts away from the mechanisms of communication. Specifically, models developed to study the cultural evolution of language ''both theoretical and computational ''often tacitly assume that linguistic signals fully specify the meaning they communicate. They imply that ignoring the fact that this is not the case in actual language use is a justified idealisation which can be made without significant consequences. In this thesis, I show that by making this idealisation, we miss out on the extensive explanatory potential of an empirically attested property of language: its pragmatic plasticity. The meaning that a signal comes to communicate in a specific context usually differs to a certain degree from its conventional meaning. This thesis (i) introduces a model of the cultural evolution of language that acknowledges and incorporates the fact that communication exhibits pragmatic plasticity and (ii) explores the explanatory potential of this fact with regard to language evolution.

The thesis falls into two parts. In the first part, I develop the model conceptually. I begin by analysing the components of extant models of general cultural evolution and discuss how models of language change and linguistic evolution map onto them. Innovative use is identified as the motor of cultural evolution. I then conceptualise the cognitive mechanisms underlying innovative language use and argue that they originate in pre-linguistic forms of ostensive-inferential communication. In a next step, the identified mechanisms are employed to provide a unified account of the two main explananda of evolutionary linguistics, the emergence of symbolism and the emergence of grammar. Finally, I discuss the implications of the presented analysis for the so-called proto-language debate. In the second part of the thesis, I propose a computational implementation of the developed conceptual model. This computational implementation allows for the simulation of the cultural emergence and evolution of symbolic communication and provides a laboratory-like environment to study individual aspects of this process. I employ such computer simulations to explore the role that pragmatic plasticity plays in the development of the expressivity, signal economy and ambiguity of emerging and evolving symbolic communication systems.

As its main contribution to the study of language evolution, this thesis shows that a model of linguistic cultural evolution that incorporates the notion of pragmatic plasticity has the potential to explain two crucial evolutionary puzzles, namely (i) how language can emerge from no language, and (ii) how language can come to exhibit the appearance of design for communication. The proposed usage-based model of language evolution bridges the evolutionary gap between no language and language by identifying ostensive-inferential communication as the continual aspect present in both stages, and by demonstrating that the cognitive mechanisms involved in ostensive-inferential communication are sufficient for the transition from one stage to the other.

The sound inventories of the world's languages show a considerable extent of symmetry. It has been postulated that this symmetry is a reflection of the human physiological, cognitive and societal factors. There have been a large number of linguistically motivated studies in order to explain the self-organization of these inventories that arguably leads to the emergence of this symmetry. A few computational models in order to explain especially the structure of the smaller vowel inventories have also been proposed in the literature. However, there is a need for a single unified computational framework for studying the self-organization of the vowel as well as other inventories of complex utterances like consonants and syllables.

In this thesis, we reformulate this problem in the light of statistical mechanics and present complex network representations of these inventories. The central objective of the thesis is to study and explain the self-organization and emergence of the consonant inventories. Nevertheless, in order to demonstrate the versatility of our modeling methodology, we further apply it to investigate and detect certain interesting properties of the vowel inventories.

Two types of networks are considered - a language-consonant bipartite network and a consonant-consonant co-occurrence network. The networks are constructed from the UCLA Phonological Segment Inventory Database (UPSID). From the systematic analysis of these networks we find that the occurrence and co-occurrence of the consonants over languages follow a well-behaved probability distribution. The co-occurrence network also exhibits a high clustering coefficient. We propose different synthetic models of network growth based on preferential attachment so as to successively match with higher accuracy the different statistical properties of the networks. Furthermore, in order to have a deeper understanding of the growth dynamics we analytically solve the models to derive expressions for the emergent degree distribution and clustering coefficient. The co-occurrence network also exhibits strong community structures and a careful inspection indicates that the driving force behind the community formation is grounded in the human articulatory and perceptual factors. In order to quantitatively validate the above principle, we introduce an information theoretic definition of this factor feature entropy and show that the natural language inventories are significantly different in terms of this quantity from the randomly generated ones. We further construct similar networks for the vowel inventories and study various interesting similarities as well as differences between them and the consonant inventories.

To summarize, this thesis shows that complex networks can be suitably used to study the self-organization of the human speech sound inventories. In this light, we deem this computational framework as a highly powerful tool in future for modeling and explaining the emergence of many other complex linguistic phenomena.

Pragmatics is the branch of linguistics that addresses the relationship between language and its external environment `` in particular the communicative context. Social evolution (or sociobiology) is the branch of the biological sciences that studies the social behaviour of organisms, particularly with respect to the ecological and evolutionary forces with which it must interact. These two disciplines thus share a natural epistemic link, one that is concerned with the relationship between behaviour and the environment. There has, however, historically been no dialogue between them. This thesis attempts to fill that void: it examines pragmatics from the perspective of social evolution theory. Chapter 1 gives a brief introduction to the two fields and their key ideas, and also discusses why an evolutionary understanding of pragmatics is crucial to the study of language origins.

In chapter 2 the vexed question of the biological function of language is discussed. Responses are given to the claims, common in the evolutionary linguistics literature, that the processes of exaptation, self organisation and cultural transmission provide alternatives to natural selection as a source of design in nature. The intuitive conclusion that the function of language is communication is provisionally supported, subject to a proper definition of communication.

Chapter 3 reviews previous definitions and consequently argues for an account predicated on the designedness of signals and responses. This definition is then used to argue that an evolutionarily coherent model of language should recognise the pragmatic realities of ostension and inference and reject the code like idealisation that is often used in its place.

Chapter 4 observes that this fits the argument that the biological function of language is communication and then addresses the key question faced by all evolved communication systems `` that of evolutionary stability. The human capacity to record and remember the past behaviour of others is seen to be critical.

Chapter 5 uses the definition of communication from chapter 3 to describe a very general model of evolved communication, and then uses the constraints of that model to argue that Relevance Theory, or at least some theory of pragmatics with a very similar logical structure, must be correct.

Chapter 6 then applies the theoretical apparatus constructed in chapters 2 to 5 to a crucial and topical issue in evolutionary linguistics: the emergence of learnt, symbolic communication. It introduces the Embodied Communication Game, an experimental tool whose basic structure is significantly informed by both social evolutionary and, in particular, pragmatic theory. The novelty of the game is that participants must find a way to communicate not just the content that they wish to convey, but also the very fact that a given behaviour is communicative in nature, and this constraint is found to fundamentally influence the type of system that emerges.

Chapter 7, which concludes the thesis, recounts and clarifies what it tells us about the origins and evolution of language, and suggests a number of possible avenues for future research.