Language Evolution and Computation Bibliography

A critical issue of distributed systems is concerned with the advertising task. Current solutions require an ex-ante agreement on a common shared language. Although such an approach is feasible from the technological point of view, it is not effective in practice. The process of managing this agreement may present social implications that make the solution difficult to achieve. Recent trends in research propose a new approach based on advertising games where the agreement on a common language is produced at run time. Nevertheless up to now such a model has been studied only through simulations with standalone platforms. Our contribution is the design and the development of the first web services oriented architecture for advertising games. Therefore we approached all the issues typical of distributed systems neglected by the simulators like asynchronous communications, denial of services, and so on. Finally we present a real world application where the architecture has been deployed to support the advertising task using an advertising game model.

Lexicon learning systems need to be concerned with more than just producing symbol usage agreement between agents, which is easy to acquire through imitation. Lexicon learners should also explicitly attempt to increase the mutual information between their symbol usages (a measure of the usefulness of the symbols for transferring information between agents). This paper argues that, although many lexicon learning algorithms presented in the literature do attempt to create highly informative symbol usages implicitly, there are good reasons to make the mutual information of symbol usages an explicit goal of the lexicon learning system. Some first steps in this direction are provided in this paper. It presents lexicon learning experiments using both purely imitative and explicitly information maximizing algorithms. The results of these experiments are used to support the thesis of this paper, that lexicon learning algorithms should explicitly attempt to produce high mutual information symbol usages.

We focus on the ``compositionality'' of semantics, a fundamental characteristic of human language, which is the ability to understand the meaning of a sentence as a combination of the meanings of words. We also pay much attention to the ``embodiment'' of a robot, which means that the robot should acquire semantics which matches its body, or sensory-motor system. The essential claim is that an embodied compositional semantic representation can be self-organized from generalized correspondences between sentences and behavioral patterns. This claim is examined and confirmed through simple experiments in which a robot generates corresponding behaviors from unlearned sentences by analogy with the correspondences between learned sentences and behaviors.

Schwa deletion is an important issue in grapheme-to-phoneme conversion for Indo- Aryan languages (IAL). In this paper, we describe a syllable minimization based algorithm for dealing with this that outperforms the existing methods in terms of efficiency and accuracy. The algorithm is motivated by the fact that deletion of schwa is a diachronic and sociolinguistic phenomenon that facilitates faster communication through syllable economy. The contribution of the paper is not just a better algorithm for schwa deletion; rather we describe here a constrained optimization based framework that can partly model the evolution of languages, and hence, can be used for solving many problems in computational linguistics that call for diachronic explanations.

A mechanism is described that enables a robotic agent to create temporal categories for conceptualizing the world. The creation of a new category is triggered when the agent is unable to temporally distinguish an event from the other events in the context using already adopted categories. This is different from most other approaches where ontological categories are defined by humans and the ontologies are fixed in advance.

We discuss which properties common-use artifacts should have to collaborate without human intervention. We conceive how devices, such as mobile phones, PDAs, and home appliances, could be seamlessly integrated to provide an ``ambient intelligence'' that responds to the users desires without requiring explicit programming or commands. While the hardware and software technology to build such systems already exists, yet there is no protocol to direct and give meaning to their interactions. We propose the first steps in the development of such a protocol, which would need to be adaptive, extensible, and open to the community, while promoting self-organization. We argue that devices, interacting through ``game-like'' moves, can learn to agree about how to communicate, with whom to cooperate, and how to delegate and coordinate specialized tasks. Like this, they may evolve distributed cognition or collective intelligence able to tackle any complex of tasks.

This paper presents an algorithm for learning the meaning of messages communicated between agents that interact while acting optimally towards a cooperative goal. Our reinforcement-learning method is based on Bayesian filtering and has been adapted for a decentralized control process. Empirical results shed light on the complexity of the learning problem, and on factors affecting the speed of convergence. Designing intelligent agents able to adapt their mutual interpretation of messages exchanged, in order to improve overall task-oriented performance, introduces an essential cognitive capability that can upgrade the current state of the art in multi-agent and human-machine systems to the next level. Learning to communicate while acting will add to the robustness and flexibility of these systems and hence to a more efficient and productive performance.

In this paper, a multi-agent computational model is proposed to simulate the coevolution of social structure and compositional protolanguage from a holistic signaling system through iterative interactions within a heterogeneous population. We implement an indirect meaning transference based on both linguistic and nonlinguistic information in communications, together with a feedback without direct meaning check. The emergent social structure, triggered by two locally selective strategies, friendship and popularity, has small-world characteristics. The influence of these selective strategies on the emergent language and the emergent social structure are discussed.

Human language serves a number of different functions, one of the most prominent being communicating about relevant features of the environment. From the point of view of the speaker, if the communicated information is advantageous for the hearer but not for the speaker, this is an altruistic use of language, and, as such, it requires an explanation of its evolution. Simon 1990 proposed an explanation of altruism in humans based on the genetically inherited docility of our species. In this paper we present artificial life simulations that apply Simon's ideas to the problem of the emergence of the altruistic use of language described above. From the point of view of evolutionary theory, the present work represents the first attempt to test Simon's ?docility? theory of altruism with agent-based computer simulations. From the point of view of language evolution, our simulations give an original explanation of (the altruistic aspect of) human language based on one of its most peculiar characteristic, namely, the fact that it is culturally transmitted.

This paper presents a computational framework for studying the influence of learning on the evolution of communication. In our model, an evolving population of learning agents is engaged in pairwise comunicative interactions. Simulation results show the genetic assimilation of trasmission behaviors as a consequence of saussurean learning.

This paper illustrates how the Zipf-Mandelbrot law can emerge in language as a result of minimising the cost of categorising sensory images. The categorisation is based on the discrimination game in which sensory stimuli are categorised at different hierarchical layers of increasing density. The discrimination game is embedded in a variant of the language game model, called the selfish game, which in turn is embedded in the framework of iterated learning. The results indicate that a tendency to communicate in general terms, which is less costly, can contribute to the emergence of the Zipf-Mandelbrot law.

This paper introduces a system that simulates the transition from the one-word stage to the two-word stage in child language production. Two-word descriptions are syntactically generated and compete against one-word descriptions from the outset. Two-word descriptions become dominant as word combinations are repeatedly recognised, forming syntactic categories; resulting in an emergent simple syntax. The system demonstrates a similar maturation as children as evidenced by phenomena such as overextensions and mismatching, and the use of one-word descriptions being replaced by two-word descriptions over time.

This paper presents an approach to simulating the evolution of language in which communication is viewed as an emerging phenomenon with both genetic and social components. A model is presented in which a population of agents is able to evolve a shared grammatical language from a purely lexical one, with critical elements of the faculty of language developed as a result of the need to navigate in and exchange information about the environment.

We present a novel connectionist model for acquiring the semantics of language through the behavioral experiences of a real robot. We focus on the ``compositionality'' of semantics, which is a fundamental characteristic of human language, namely, the fact that we can understand the meaning of a sentence as a combination of the meanings of words. The essential claim is that a compositional semantic representation can be self-organized by generalizing correspondences between sentences and behavioral patterns. This claim is examined and confirmed through simple experiments in which a robot generates corresponding behaviors from unlearned sentences by analogy with the correspondences between learned sentences and behaviors.

Communication is often required for coordination of collective behaviours. Social insects like ants, termites or bees make use of different forms of communication, which can be roughly classified in three classes: indirect (stigmergic) communication, direct interaction and direct communication. The use of stigmergic communication is predominant in social insects (e.g., the pheromone trails in ants), but also direct interactions (e.g., antennation in ants) and direct communication can be observed (e.g., the waggle dance of honey bee workers). Direct communication may be beneficial when a fast reaction is expected, as for instance, when a danger is detected and countermeasures must be taken. This is the case of hole avoidance, the task studied in this paper: a group of selfassembled robots - called swarm-bot - coordinately explores an arena containing holes, avoiding to fall into them. In particular, we study the use of direct communication in order to achieve a reaction to the detection of a hole faster than with the sole use of direct interactions through physical links. We rely on artificial evolution for the synthesis of neural network controllers, showing that evolving behaviours that make use of direct communication is more effective than exploiting direct interactions only.

A new sign language has been created by deaf Nicaraguans over the past 25 years, providing an opportunity to observe the inception of universal hallmarks of language. We found that in their initial creation of the language, children analyzed complex events into basic elements and sequenced these elements into hierarchically structured expressions according to principles not observed in gestures accompanying speech in the surrounding language. Successive cohorts of learners extended this procedure, transforming Nicaraguan signing from its early gestural form into a linguistic system. We propose that this early segmentation and recombination reflect mechanisms with which children learn, and thereby perpetuate, language. Thus, children naturally possess learning abilities capable of giving language its fundamental structure.

Nicaraguan Sign Language (NSL), developed by deaf children in Managua over the past 35 years, has provided unprecedented insights into the innateness of language. In his Perspective, Siegal discusses the latest study of three age groups of NSL signers (Senghas et al.), which reveals that segmentation and sequencing, considered core properties of all languages, are clearly present in NSL.

Traditional language learning theory explores an idealized interaction between a teacher and a learner. The teacher provides sentences from a language, while the learner has to infer the underlying grammar. Here, we study a new approach by considering a population of individuals that learn from each other. There is no designated teacher. We are inspired by the observation that children grow up to speak the language of their peers, not of their parents. Our goal is to characterize learning strategies that generate ``linguistic coherence,'' which means that most individuals use the same language. We model the resulting learning dynamics as a random walk of a population on a graph. Each vertex represents a candidate language. We find that a simple strategy using a certain aspiration level with the principle of win-stay, lose-shift does extremely well: stay with your current language, if at least three others use that language; otherwise, shift to an adjacent language on the graph. This strategy guarantees linguistic coherence on all nearly regular graphs, in the relevant limit where the number of candidate languages is much greater than the population size. Moreover, for many graphs, it is sufficient to have an aspiration level demanding only two other individuals to use the same language.

In a recent article Mitchener and Nowak present a model of the evolutionary dynamics of language. The model exhibits regular and chaotic oscillations in changes to the proportions of grammars spoken in a population over the course of evolution. These oscillations are within the purview of evolutionary game theory, but they suggest the lack of an evolutionarily stable strategy. Implications for self-organization across scales of adaptation are discussed.

Recent demonstrations of statistical learning in infants have reinvigorated the innateness versus learning debate in language acquisition. This article addresses these issues from both computational and developmental perspectives. First, I argue that statistical learning using transitional probabilities cannot reliably segment words when scaled to a realistic setting (e.g. childdirected English). To be successful, it must be constrained by knowledge of phonological structure. Then, turning to the bona fide theory of innateness - the Principles and Parameters framework - I argue that a full explanation of children's grammar development must abandon the domain-specific learning model of triggering, in favor of probabilistic learning mechanisms that might be domain-general but nevertheless operate in the domain-specific space of syntactic parameters.

Replicator-mutator equation is used to describe the dynamics of complex adaptive systems in population genetics, biochemistry and models of language learning. We study 'localized', or 'coherent', solutions, which are especially relevant in the context of learning and correspond to the existence of a predominant language in the population. There is a coherence threshold for learning fidelity, above which coherent communication can be maintained. We prove the following surprising universality property of coherence threshold: for typical realizations of random coefficients in the fitness matrix, the value of the coherence threshold does not depend on the size of the system.

Human language is unique among the communication systems of the natural world. The vocabulary of human language is unique in being both culturally-transmitted and symbolic. In this paper I present an investigation into the factors involved in the evolution of such vocabulary systems. I investigate both the cultural evolution of vocabulary systems and the biological evolution of learning rules for vocabulary acquisition.

Firstly, vocabularies are shown to evolve on a cultural time-scale so as to fit the expectations of learners -- a population's vocabulary adapts to the biases of the learners in that population. A learning bias in favour of one-to-one mappings between meanings and words leads to the cultural evolution of communicativelyoptimal vocabulary systems, even in the absence of any explicit pressure for communication. Furthermore, the pressure to conform to the biases of learners is shown to outweigh natural selection acting on cultural transmission. Human language learners appear to bring a one-to-one bias to the acquisition of vocabulary systems. The functionality of human vocabulary may therefore be a consequence of the biases of human language learners.

Secondly, the evolutionary stability of genetically-transmitted vocabulary learning biases is investigated using both static and dynamic models. A one-to-one learning bias, which leads to the cultural evolution of optimal communication, is shown to be evolutionarily stable. However, the evolution de novo of this bias is complicated by the cumulative nature of the cultural evolution of vocabulary systems. This suggests that the biases of human language learners may not have evolved specifically and exclusively for the acquisition of communicatively-functional vocabulary.

Infants learn language with remarkable speed, but how they do it remains a mystery. New data show that infants use computational strategies to detect the statistical and prosodic patterns in language input, and that this leads to the discovery of phonemes and words. Social interaction with another human being affects speech learning in a way that resembles communicative learning in songbirds. The brain's commitment to the statistical and prosodic patterns that are experienced early in life might help to explain the long-standing puzzle of why infants are better language learners than adults. Successful learning by infants, as well as constraints on that learning, are changing theories of language acquisition.

We present the high-level language of relational growth grammars (RGGs) as a formalism designed for the specification of ALife models. RGGs can be seen as an extension of the well-known parametric Lindenmayer systems and contain rule-based, procedural, and object-oriented features. They are defined as rewriting systems operating on graphs with the edges coming from a set of user-defined relations, whereas the nodes can be associated with objects. We demonstrate their ability to represent genes, regulatory networks of metabolites, and morphologically structured organisms, as well as developmental aspects of these entities, in a common formal framework. Mutation, crossing over, selection, and the dynamics of a network of gene regulation can all be represented with simple graph rewriting rules. This is demonstrated in some detail on the classical example of Dawkins' biomorphs and the ABC model of flower morphogenesis: other applications are briefly sketched. An interactive program was implemented, enabling the execution of the formalism and the visualization of the results.

Different theories have been proposed for explaining the evolution of language. One of this maintains that gestural communication has been the precursor of human speech. Here we present a series of neurophysiological evidences that support this hypothesis. Communication by gestures, defined as the capacity to emit and recognize meaningful actions, may have originated in the monkey motor cortex from a neural system whose basic function was action understanding. This system is made by neurons of monkey’s area F5, named mirror neurons, activated by both execution and observation of goal-related actions. Recently, two new categories of mirror neurons have been described. Neurons of one category respond to the sound of an action, neurons of the other category respond to the observation of mouth ingestive and communicative actions. The properties of these neurons indicate that monkey's area F5 possesses the basic neural mechanisms for associating gestures and meaningful sounds as a pre-adaptation for the later emergence of articulated speech. The homology and the functional similarities between monkey area F5 and Broca’s area support this evolutionary scenario.

In the last ten years, researchers have accepted that language is the outcome of normal evolutionary processes. This has led to a flurry of new work, resulting in some important steps forward in our understanding of language origin. This paper outlines the highlights. The location of proto-humans has been confirmed as Africa, and African hominids are more widely scattered than was previously assumed. Some probably moved out of Africa earlier than was once thought likely. Evolutionary theory has also been explored in more depth. The probable date of language origin has moved earlier, and the precursors of language have been examined, resulting in a more sophisticated understanding of symbolic communication. Mirror neurons have been proposed as relevant to the neurological underpinnings of mind-reading, which may underlie the naming insight. A proto-language stage probably preceded full language, and this proto-language may have involved several layers, which are still visible in language today.

Human language is qualitatively different from animal communication systems in at least two separate ways. Human languages contain tens of thousands of arbitrary learned symbols (mainly words). No other animal communication system involves learning the component symbolic elements afresh in each individual's lifetime, and certainly not in such vast numbers. Human language also has complex compositional syntax. The meanings of our sentences are composed from the meanings of the constituent parts (e.g. the words). This is obvious to us, but no other animal communication system (with honeybees as an odd but distracting exception) puts messages together in this way. A recent theoretical claim that the sole distinguishing feature of human language is recursion is discussed, and related to these features of learned symbols and compositional syntax. It is argued that recursive thought could have existed in prelinguistic hominids, and that the key step to language was the innovative disposition to learn massive numbers of arbitrary symbols

How do communities establish shared communication systems? The Common Knowledge view assumes that symbolic conventions develop through the accumulation of common knowledge regarding communication practices among the members of a community. In contrast with this view, it is proposed that coordinated communication emerges a by-product of local interactions among dyads. A set of multi-agent computer simulations show that a population of 'egocentric' agents can establish and maintain symbolic conventions without common knowledge. In the simulations, convergence to a single conventional system was most likely and most efficient when agents updated their behavior on the basis of local rather than global, system-level information. The massive feedback and parallelism present in the simulations gave rise to phenomena that are often assumed to result from complex strategic processing on the part of individual agents. The implications of these findings for the development of theories of language use are discussed.

The paper presents a computational model of language in which linguistic abilities evolve in organisms that interact with an environment. Each individual's behavior is controlled by a neural network and we study the consequences in the network's internal functional organization of learning to process different classes of words. Agents are selected for reproduction according to their ability to manipulate objects and to understand nouns (objects' names) and verbs (manipulation tasks). The weights of the agents' neural networks are evolved using a genetic algorithm. Synthetic brain imaging techniques are then used to examine the functional organization of the neural networks. Results show that nouns produce more integrated neural activity in the sensory-processing hidden layer, while verbs produce more integrated synaptic activity in the layer where sensory information is integrated with proprioceptive input. Such findings are qualitatively compared with human brain imaging data that indicate that nouns activate more the posterior areas of the brain related to sensory and associative processing, while verbs activate more the anterior motor areas.

This paper presents the hypothesis that linguistic capacity evolved through the action of natural selection as an instrument which increased the efficiency of the cultural transmission system of early hominids. We suggest that during the early stages of hominization, hominid social learning, based on indirect social learning mechanisms and true imitation, came to constitute cumulative cultural transmission based on true imitation and the approval or disapproval of the learned behaviour of offspring. A key factor for this transformation was the development of a conceptual capacity for categorizing learned behaviour in value terms - positive or negative, good or bad. We believe that some hominids developed this capacity for categorizing behaviour, and such an ability allowed them to approve or disapprove of their offsprings- learned behaviour. With such an ability, hominids were favoured, as they could transmit to their offspring all their behavioural experience about what can and cannot be done. This capacity triggered a cultural transmission system similar to the human one, though pre-linguistic. We suggest that the adaptive advantage provided by this new system of social learning generated a selection pressure in favour of the development of a linguistic capacity allowing children to better understand the new kind of evaluative information received from parents.

In an effort to better guide research into self-configuring wireless sensor networks, we discuss a technical definition of the term self-organization. We define a self-organizing system as one where a collection of units coordinate with each other to form a system that adapts to achieve a goal more efficiently. We then lay out some conditions that must hold for a system to meet this definition and discuss some examples of self-organizing systems. Finally, we explore some of the ways this definition applies to wireless sensor networks.

Among the several findings deriving from the application of complex network formalism to the investigation of natural phenomena, the fact that linguistic constructions follow power laws presents special interest for its potential implications for psychology and brain science. By corresponding to one of the most essentially human manifestations, such language-related properties suggest that similar dynamics may also be inherent to the brain areas related to language and associative memory, and perhaps even consciousness. The present work reports a preliminary experimental investigation aimed at characterizing and modeling the flow of sequentially induced associations between words from the English language in terms of complex networks. The data is produced through a psychophysical experiment where a word is presented to the subject, who is requested to associate another word. Complex network and graph theory formalism and measurements are applied in order to characterize the experimental data. Several interesting results are identified, including the characterization of attraction basins, association asymmetries, context biasing, as well as a possible power-law underlying word associations, which could be explained by the appearance of strange loops along the hierarchical structure underlying word categories.

This study examines certain mechanisms underlying the evolution of language and science - including mathematics - using concepts developed in the field of biological evolution. Developmental processes are particularly emphasized. Analysis of developmental processes, processes such as human embryonic development, children's verbal development, and adolescents' scientific conceptual development reveals the unifying principle referred to as 'condensation' - the successive shortening of developmental stages. The mechanism of condensation is coupled to the rate of evolutionary change.

The analysis examines the applicability of the concept of the meme. Regarding the evolution of language, we suggest a cooperative combination of genetic and memetic replication; while early on in the evolution of science only memetic replication is envisaged.

We study the Euclidean distance between syntactically linked words in sentences. The average distance is significantly small and is a very slowly growing function of sentence length. We consider two nonexcluding hypotheses: (a) the average distance is minimized and (b) the average distance is constrained. Support for (a) comes from the significantly small average distance real sentences achieve. The strength of the minimization hypothesis decreases with the length of the sentence. Support for (b) comes from the very slow growth of the average distance versus sentence length. Furthermore, (b) predicts, under ideal conditions, an exponential distribution of the distance between linked words, a trend that can be identified in real sentences.

Many languages are spoken on Earth. Despite their diversity, many robust language universals are known to exist. All languages share syntax, i.e., the ability of combining words for forming sentences. The origin of such traits is an issue of open debate. By using recent developments from the statistical physics of complex networks, we show that different syntactic dependency networks (from Czech, German, and Romanian) share many nontrivial statistical patterns such as the small world phenomenon, scaling in the distribution of degrees, and disassortative mixing. Such previously unreported features of syntax organization are not a trivial consequence of the structure of sentences, but an emergent trait at the global scale.

Zipf's law asserts that in all natural languages the frequency of a word is inversely proportional to its rank. The significance, if any, of this result for language remains a mystery. Here we examine a null hypothesis for the distribution of word frequencies, a so-called discourse-triggered word choice model, which is based on the assumption that the more a word is used, the more likely it is to be used again. We argue that this model is equivalent to the neutral infinite-alleles model of population genetics and so the degeneracy of the different words composing a sample of text is given by the celebrated Ewens sampling formula [

Theor. Pop. Biol. 3, 87 (1972)

], which we show to produce an exponential distribution of word frequencies.

The order of 'noun and adposition' is an important parameter of word ordering rules in the world's languages. The seven parameters, 'adverb and verb' and others, depend strongly on the 'noun and adposition'. Japanese as well as Korean, Tamil and several other languages seem to have a stable structure of word ordering rules, while Thai and other languages, which have the opposite word ordering rules to Japanese, are also stable in structure. It seems therefore that each language in the world fluctuates between these two structures like the Ising model for finite lattice.

We consider the problem of linguistic agents that communicate with each other about a shared world. We develop a formal notion of a language as a set of probabilistic associations between form (lexical or syntactic) and meaning (semantic) that has general applicability. Using this notion, we define a natural measure of the mutual intelligibility, F(L,L'), between two agents, one using the language L and the other using L'. We then proceed to investigate three important questions within this framework: (1) Given a language L, what language L' maximizes mutual intelligibility with L? We find surprisingly that L' need not be the same as L and we present algorithms for approximating L' arbitrarily well. (2) How can one learn to optimally communicate with a user of language L when L is unknown at the outset and the learner is allowed a finite number of linguistic interactions with the user of L? We describe possible algorithms and calculate explicit bounds on the number of interactions needed. (3) Consider a population of linguistic agents that learn from each other and evolve over time. Will the community converge to a shared language and what is the nature of such a language? We characterize the evolutionarily stable states of a population of linguistic agents in a game-theoretic setting. Our analysis has significance for a number of areas in natural and artificial communication where one studies the design, learning, and evolution of linguistic communication systems.

What constitutes linguistic evidence for Universal Grammar (UG)? The principal approach to this question equates UG on the one hand with language universals on the other. Parsimonious and general characterizations of linguistic variation are assumed to uncover features of UG. This paper reviews a recently developed evolutionary approach to language that casts doubt on this assumption: the Iterated Learning Model (ILM). We treat UG as a model of our prior learning bias, and consider how languages may adapt in response to this bias. By dealing directly with populations of linguistic agents, the ILM allows us to study the adaptive landscape that particular learning biases result in. The key result from this work is that the relationship between UG and language structure is non-trivial.

In this paper we present a self-organizing neural network model of early lexical development called DevLex. The network consists of two self-organizing maps (a growing semantic map and a growing phonological map) that are connected via associative links trained by Hebbian learning. The model captures a number of important phenomena that occur in early lexical acquisition by children, as it allows for the representation of a dynamically changing linguistic environment in language learning. In our simulations, DevLex develops topographically organized representations for linguistic categories over time, models lexical confusion as a function of word density and semantic similarity, and shows age-of-acquisition effects in the course of learning a growing lexicon. These results match up with patterns from empirical research on lexical development, and have significant implications for models of language acquisition based on self-organizing neural networks.

Human language is a complex and expressive communication system. Children spontaneously develop a native language from speech they hear in their community. Languages change dramatically and unpredictably by accumulating small changes over time and by interacting with other languages. This paper describes a mathematical model illustrating language change. Children learn their parents' language imperfectly, and in the case presented here, the result is a simulated population that maintains an ever-changing mixture of grammars. This research is part of a growing attempt to use mathematical models to better understand the social and biological history of language.

Since the late 1980s, language endangerment and death have been discussed as if the phenomena had no connection at all with language birth. More recently the phenomena have been associated almost exclusively with the intense and pervasive economic globalization of same period, a process that some authors have reduced too easily to the McDonaldization phenomenon. Moreover, the relation of globalization to different forms of colonization has been poorly articulated. As a matter of fact, little of the longer history of population movements and contacts since the dawn of agriculture has been invoked in the literature on language endangerment to give some broader perspective on the mechanisms of language birth and death and on the ecological factors that bear on how they proceed. This review aims to remedy these shortcomings in our scholarship.

In order to learn grammar from a finite amount of evidence, children must begin with in-built expectations of what is grammatical. They clearly are not born, however, with fully developed grammars. Thus early language development involves refinement of the grammar hypothesis until a target grammar is learnt. Here we address the question of how much evidence is required for this refinement process, by considering two standard learning algorithms and a third algorithm which is presumably as efficient as a child for some value of its memory capacity. We reformulate this algorithm in the context of Chomsky's 'principles and parameters' and show that it is possible to bound the amount of evidence required to almost certainly speak almost grammatically.

We consider a model introduced recently [Nature 424(2003)900], for describing competition between two languages, which in typical situations predicts the extinction of one of them. We generalize it by introducing a spatial dependence in terms of a reaction-diffusion equation. We show that in this generalized model both languages can survive, each mostly concentrated in a different geographical area.

Music functions, at least in part, to convey certain structures to the listener via a surface of notes. For communication to occur successfully, the structures must be recoverable from the surface. I argue that this consideration has been an important factor in the shaping of musical styles, and sheds light on a number of phenomena: the greater degree of syncopation and lower degree of rubato in traditional African music and rock versus common-practice music; the extensive use of rubato in pieces with consistent repeated patterns (e.g., much Romantic piano music); the rise of swing tempo and the higher degree of syncopation in jazz as opposed to ragtime; and the greater variety of chord-tones and lower tolerance for chordal inversion in jazz as opposed to common-practice music.

In this paper, the basic words introduced by Swadesh(1952,1955) are used to uncover information about language evolution. In section 2, based on Chen (1996), Swadesh's 200-word list can be split into two sub-groups; the first 100 words, called high rank, are more stable and loan-resistant than the second 100, called low rank. More retentions tend to be present among the high rank words while borrowed elements tend to occur more frequently among the low rank words. This feature can be used to distinguish retentions and borrowings when the evolution of a language has been blurred by language contact, such as the transmission of Middle Chinese entering tones into Pekinese. In section 3, we argue that Swadesh (1955)'s 100 basic words (high rank) are better for sub-grouping Chinese dialects, compared with Dolgopolsky (1964)'s 15-word list and Yakhontov' 35-word list.

At five weeks old, my normally happy first child became a fussy baby. About midday one day, he began to cry. After trying all of my standard calming techniques to no avail, in frustration I finally told him, loudly," STOP!" This stopped the crying, but I felt so guilty about yelling at ...

Form: The repertoire of speech sounds used in human language is extraordinarily complex. It relies on an articulatory apparatus which needs to be controlled very fast and at a very fine-grained level. It requires the real-time processing of structured sounds despite noise and ...

Language is transmitted from one generation to the next via learning by individuals. By taking this point of view one is able to link the linguistic behavior of successive generations and therefore study how language evolves over generational time scales. We provide a brief overview of this approach to the study of language evolution, its formalization as a dynamical system, and the analogical connections to the methodological principles of evolutionary biology. We show how the interplay between learning and evolution can be quite subtle and how phase transitions arise in many such models of language evolution. Such phase transitions may provide a suitable theoretical construct with which explanations for rapid language change or evolution may be given. Some illustrative examples are provided.

The study of language evolution has revitalized recently due to converging interests from many disciplines. Computational modeling is one such fruitful area. Various aspects of language evolution have been studied using mathematical modeling and simulation. In this paper we discuss several computational studies in language change and language emergence.

This book provides a current and interdisciplinary overview of work on the biology of language - what is sometimes called the ``biolinguistic approach.'' A wide range of areas are investigated and reviewed by specialists: the micro-parametric theory of syntax, models of language acquisition and historical change, dynamical systems in language, genetics of populations, pragmatics of discourse, language neurology, genetic disorders of language, sign language, and evolution of language.

TABLE OF CONTENTS
Variation in Typology, Acquisition and Change.
Antisymmetry and Japanese. (R. Kayne). Toward a Theory of Language Growth. (C. Yang). Phase Transitions in Language Evolution. (P. Niyogi).

Variation in Genetics and Domain Specificity.
Genetic Differences and Language Affinities. (I. Dupanloup). Beyond Narrow Syntax. (S. Avrutin). Evidence for and Implications of a Domain-Specific Grammatical Deficit. (H.K.J. van der Lely).

Neurological Variation and Language Emergence.
The Representation of Grammatical Knowledge in the Brain. (A. Caramazza, K. Shapiro). Variation in Broca's Region: Preliminary Cross-Methodological Comparisons. (Y. Grodzinsky). Language Emergence in a Language-Ready Brain: Acquisition. (J. Kegl).

Variation in Development Genetics and Language Disorders.
Lenneberg's Dream: Learning, Normal Language Development and Specific Language Impairment. (K. Wexler). Exploring the Phenotype of Specific Language Impairment: a Look at Grammatical Variability. (L. Leonard). The Investigation of Genetic Dysphasia. (M. Gopnik).

Unification of Linguistics into the Natural Sciences.
Unification in Biolinguistics. (L. Jenkins). The Immune Syntax: The Evolution of the Language Virus. (M. Piattelli-Palmarini, J. Uriagereka). Language and Mind: Current Thoughts on Ancient Problems. (N. Chomsky).

This thesis investigates language acquisition and evolution, using the methodologies of Bayesian inference and expression-induction modelling, making specific reference to colour term typology, and syntactic acquisition. In order to test Berlin and Kay's (1969) hypothesis that the typological patterns observed in basic colour term systems are produced by a process of cultural evolution under the influence of universal aspects of human neurophysiology, an expression-induction model was created. Ten artificial people were simulated, each of which was a computational agent. These people could learn colour term denotations by generalizing from examples using Bayesian inference, and the resulting denotations had the prototype properties characteristic of basic colour terms. Conversations between these people, in which they learned from one-another, were simulated over several generations, and the languages emerging at the end of each simulation were investigated. The proportion of colour terms of each type correlated closely with the equivalent frequencies found in the World Colour Survey, and most of the emergent languages could be placed on one of the evolutionary trajectories proposed by Kay and Maffi (1999). The simulation therefore demonstrates how typological patterns can emerge as a result of learning biases acting over a period of time.

Further work applied the minimum description length form of Bayesian inference to modelling syntactic acquisition. The particular problem investigated was the acquisition of the dative alternation in English. This alternation presents a learnability paradox, because only some verbs alternate, but children typically do not receive reliable evidence indicating which verbs do not participate in the alternation (Pinker, 1989). The model presented in this thesis took note of the frequency with which each verb occurred in each subcategorization, and so was able to infer which subcategorizations were conspicuously absent, and so presumably ungrammatical. Crucially, it also incorporated a measure of grammar complexity, and a preference for simpler grammars, so that more general grammars would be learned unless there was sufficient evidence to support the incorporation of some restriction. The model was able to learn the correct subcategorizations for both alternating and non-alternating verbs, and could generalise to allow novel verbs to appear in both constructions. When less data was observed, it also overgeneralized the alternation, which is a behaviour characteristic of children when they are learning verb subcategorizations. These results demonstrate that the dative alternation is learnable, and therefore that universal grammar may not be necessary to account for syntactic acquisition. Overall, these results suggest that the forms of languages may be determined to a much greater extent by learning, and by cumulative historical changes, than would be expected if the universal grammar hypothesis were correct.

This thesis proposes a framework adopting the self-organization theory for the study of language evolution. Self-organization explains collective behaviors and evolution with the observation that the patterns at the global level in a complex system are often properties spontaneously emergent from the numerous local interactions among the individual components, and they cannot be understood by only examining the individual components.

Language can be viewed as such emergent properties instead of products from some innate blueprint in humans. We highlight the importance of recognizing language at two distinctive but inter-dependent levels of existence, i.e. in the idiolect and in the communal language, and a self-organizing process existing at each of the two levels. It is necessary to clarify what phenomena are properties of the idiolects, and what properties are the collective behaviors at the population level.

In linguistics, however, very often an abstract language system is taken as the object of analysis. This level of analysis disregards the distinction between idiolect and communal language, and neglects the heterogeneous nature of language at both levels. As a consequence, explanations for observed patterns based on this abstract level of analysis are often inadequate. However, this is a necessary step for linguists to identify interesting phenomena in the first place. At this abstract level of analysis, the self-organization framework can also be applied. It is assumed that the abstract language system self-organizes. A study on homophony in languages is taken as an example to illustrate the analysis at this level. It is shown that the existence of homophony reflects several self-organization characteristics in a dynamic process of language evolution, such as the predictable degree of homophony, the disyllabification in Chinese dialects, the differentiation of homophone pairs in grammatical class.

We are further interested in how the self-organization is implemented. To answer this question, we need to look into the idiolects in this self-organizing process, to know how the idiolects are formed and affect each other. Language change provides an informative window in addressing these issues. Language change is the result of the collective behaviors of idiolects, even as it affects the idiolects. The heterogeneity among idiolects is exposed to the greatest extent in on-going changes.

An on-going sound change in Cantonese is taken as a case study to scrutinize the heterogeneity in the self-organizing processes. The fieldwork data reveal a large degree of variation both in the population (VT-I) and in the set of words (VT-II). Another type of variation (VT-III) is highlighted, that is, a word may also show variation within one single speaker. But this VT-III within speakers only exists in a proportion, but not all, of the words subject to the change. Also we find that if a speaker has some words consistently in the unchanged state and some words in the changed state, then this speaker must have some other words in the variation state. Most speakers show the existence of VT-III, but they vary in degree. The observed individual differences in the degree of VT-III suggest that the large heterogeneity may be not only accounted for by the variability of linguistic input, but also by individuals' different learning styles. We hypothesize two types of lexical learning styles, i.e. probabilistic and categorical learning. These differences in learning styles suggest that when we examine the agent's internal properties in the self-organization framework, it is not only necessary to examine the commonalities among agents, but also the differences among them.

In addition to empirical studies, this thesis employs computational modeling as a major tool for investigation, as modeling provides effective ways to test hypotheses beyond empirical studies, and suggests new questions. After a brief review of the modeling studies in the field, some models developed in this thesis for language origin and language change are reported.

The first model is to simulate the emergence of a consistent vocabulary from a set of random mappings between meanings and forms. It emphasizes the importance of implementing the actual process of interaction among agents, and the cumulative effect on agents' linguistic behaviors. The model suggests that the Saussurean sign with identical speaking and listening mappings may not be a biological predisposition from natural selection, but rather a result from the process of language learning and use. The process exhibits a phase transition from a long period of small oscillation to an abrupt convergence. Such phase transition is often observed in self-organizing systems.

The second model simulates language change as innovation diffusion, and examines the effects of various factors, including some concerning properties of agents and some affecting agents' interactions. By comparing the outcome under different conditions, the model illustrates the importance of incorporating realistic assumptions, such as finite population size, age-dependent propensity to change, different learning environment in a social network, etc. The model compares the dynamics of language change in different types of network structures and shows that in non-regular networks, the rate of innovation diffusion increases little as population size increases. The model also tests the effect of the two types of hypothesized learning styles, and shows that in a population with the presence of probabilistic learners, an innovation with a small advantage will easily spread into the population and lead to a change. This may explain why language changes are so frequent.

This thesis demonstrates that both empirical and modeling studies on language evolution can greatly benefit from adopting a self-organization framework. The convergence and interplay of the two lines of exploration, i.e. biological bases in agents and the long term effect of interactions among them, should bring us a deeper understanding of how language has evolved and is evolving.

Generative models of phonology account for output patterns through a complex grammar applied over a minimal lexicon. In contrast, many natural complex patterns result from the gradual accumulation of structure through repeated local interactions. In this dissertation I present results of simulations supporting the proposal that some phonological patterns can be accounted for through self-organization within an analogically structured lexicon, in response to forcing from external biases. In Chapter 1, I show that patterns accounted for by the Optimality-Theoretic principles of constraint dominance and strict constraint dominance can be shown to spontaneously arise in analogically-structured systems, driven by competition between leveling pressures within the lexicon and differentiating pressures from lexicon- external performance biases.

Phonological systems exhibit `constrained contrast' in two distinct ways: first, phonologies exhibit only a subset of cross-linguistically attested contrasts, formed from a subset of possible features in combination. Second, crosslinguistically infrequent elements also tend to occur less frequently in a language that does have them. In Chapter 2, I present evidence that both of these patterns can be accounted for diachronically through indirect selection over phonetic variants, given the assumptions that, 1) lexical categories are richly specified, 2) a perceived utterance updates the content of a lexical category only if it is identified as an example of that lexical category, and 3) lexical categories can influence each others' production in proportion to phonological similarity.

When a simulated speaker/hearer pair alternately communicate their lexicons to each other under these conditions their lexicons converge. Further, when an output is too close to multiple categories, it is less likely be consistently categorized, with the result that it has less influence on the evolution of the pairs' lexicons, resulting in pressure on lexical categories to remain contrastive. W hen biases against certain features or feature combinations are introduced, the pairs' lexicons evolve to avoid as many of these `less-fit' elements as possible. However, when avoidance of all marked elements would result in insufficient contrast, the lexicons evolve to utilize a subset of less-fit elements, but at a lower frequency than fitter elements.

Since Hinton & Nowlan published their seminal paper (Hinton & Nowlan 1987), the neglected evolutionary process of the Baldwin effect has been widely acknowledged. Especially in the field of language evolution, the Baldwin effect (Baldwin 1896d, Simpson 1953) has been expected to salvage the long-lasting deadlocked situation of modern linguistics: i.e., it may shed light on the relationship between environment and innateness in the formation of language.

However, as intense research of this evolutionary theory goes on, certain robust difficulties have become apparent. One example is genotype-phenotype correlation. By computer simulations, both Yamauchi (1999, 2001) and Mayley (1996b) show that for the Baldwin effect to work legitimately, correlation between genotypes and phenotypes is the most essential underpinning. This is due to the fact that this type of the Baldwin effect adopts as its core mechanism Waddington's (1975) `genetic assimilation'. In this mechanism, phenocopies have to be genetically closer to the innately predisposed genotype. Unfortunately this is an overly na{umlaut}yssumption for the theory of language evolution. As a highly complex cognitive ability, the possibility that this type of genotype-phenotype correlation exists in the domain of linguistic ability is vanishingly small.

In this thesis, we develop a new type of mechanism, called `Baldwinian Niche Construction (BNC), that has a rich explanatory power and can potentially overcome this bewildering problem of the Baldwin effect. BNC is based on the theory of niche construction that has been developed by Odling-Smee et al. (2003). The incorporation of the theory into the Baldwin effect was first suggested by Deacon (1997) and briefly introduced by Godfrey-Smith (2003). However, its formulation is yet incomplete.

In the thesis, first, we review the studies of the Baldwin effect in both biology and the study of language evolution. Then the theory of BNC is more rigorously developed. Linguistic communication has an intrinsic property that is fundamentally described in the theory of niche construction. This naturally leads us to the theoretical necessity of BNC in language evolution. By creating a new linguistic niche, learning discloses a previously hidden genetic variance on which the Baldwin `canalizing' effect can take place. It requires no genetic modification in a given genepool. There is even no need that genes responsible for learning occupy the same loci as genes for the innate linguistic knowledge. These and other aspects of BNC are presented with some results from computer simulations.