For a long time, the accepted explanatory model of language acquisition was the so-called Principles and Parameters framework (P&P). P&P seemingly provides an elegant answer to the central puzzle of generative linguistics: How do children acquire their native language given the limited time and input resources available to them? Yet P&P tells a story that is evolutionarily implausible, and for this reason it has since been abandoned. I argue that this is an unwarranted move, and that it could and should (...) be avoided by reassessing the epistemic status of P&P. In particular, I argue that contrary to extant accounts, P&P ought to be retrospectively construed as a highly idealized (toy) model of language acquisition. The proposed reinterpretation is vindicated, I argue, insofar as it paves the way for a reconciliation of the two central explanatory challenges of modern generative linguistics. (shrink)

One of the classic debates in cognitive science is between nativism and empiricism about the development of psychological capacities. In principle, the debate is empirical. However, in practice nativist hypotheses have also been challenged for relying on an ill-defined, or even unscientific, notion of innateness as that which is “not learned”. Here this minimal conception of innateness is defended on four fronts. First, it is argued that the minimal conception is crucial to understanding the nativism-empiricism debate, when properly construed; Second, (...) various objections to the minimal conception—that it risks overgeneralization, lacks an account of learning, frustrates genuine explanations of psychological development, and fails to unify different notions of innateness across the sciences—are rebutted. Third, it is argued that the minimal conception avoids the shortcomings of primitivism, the prominent view that innate capacities are those that are not acquired via a psychological process in development. And fourth, the minimal conception undermines some attempts to identify innateness with a natural kind. So in short, we have little reason to reject, and good reason to accept, the minimal conception of innateness in cognitive science. (shrink)

This article surveys the philosophical literature on theoretical linguistics. The focus of the paper is centred around the major debates in the philosophy of linguistics, past and present, with specific relation to how they connect to the philosophy of science. Specific issues such as scientific realism in linguistics, the scientific status of grammars, the methodological underpinnings of formal semantics, and the integration of linguistics into the larger cognitive sciences form the crux of the discussion.

One of the great successes of the application of generalized quantifiers to natural language has been the ability to formulate robust semantic universals. When such a universal is attested, the question arises as to the source of the universal. In this paper, we explore the hypothesis that many semantic universals arise because expressions satisfying the universal are easier to learn than those that do not. While the idea that learnability explains universals is not new, explicit accounts of learning that can (...) make good on this hypothesis are few and far between. We propose a model of learning — back-propagation through a recurrent neural network — which can make good on this promise. In particular, we discuss the universals of monotonicity, quantity, and conservativity and perform computational experiments of training such a network to learn to verify quantifiers. Our results are able to explain monotonicity and quantity quite well. We suggest that conservativity may have a different source than the other universals. (shrink)

This paper defends the view that the Faculty of Language is compositional, i.e., that it computes the meaning of complex expressions from the meanings of their immediate constituents and their structure. I fargue that compositionality and other competing constraints on the way in which the Faculty of Language computes the meanings of complex expressions should be understood as hypotheses about innate constraints of the Faculty of Language. I then argue that, unlike compositionality, most of the currently available non-compositional constraints predict (...) incorrect patterns of early linguistic development. This supports the view that the Faculty of Language is com- positional. More generally, this paper presents a way of framing the compositionality debate (by focusing on its implications for language acquisition) that can lead to its even- tual resolution, so it will hopefully also interest theorists who disagree with its main conclusion. (shrink)

This project provides a detailed examination and critique of current philosophical, linguistic, and cognitive accounts of first language acquisition. In particular, I focus on the concept of "innate" and how it is embraced, marginally utilized, or abandoned altogether in efforts to describe the way that a child comes to be a competent user of a language. A central question that naturally falls out of this general inquiry is therefore what exactly is supposed to be "innate," according to various theories? Philosophically, (...) the theory of innate ideas put forth to explain human learning has existed for centuries and hence, this thesis as it relates to language is discussed. The revival of nativism by linguists like Chomsky is thus a central theme of the first chapter. Universal Grammar and the various arguments for it are closely scrutinized, and I close this chapter with what I take to be the commitments of linguistic nativism, how its proponents conceive of "innate," and several possible objections to the arguments they put forth. Just as the theory of innate ideas has had its contesters throughout the history of philosophy, so too have linguists and cognitive scientists rejected Universal Grammar and other forms of linguistic nativism. Thus, the second chapter presents several of these alternative explanations of language acquisition. Namely, I divide the chapter into three sections, Usage-Based Linguistics, Emergentism, and Sociolinguistic Acquisition, as it is my suggestion that most of the anti-UG attacks are levied from one of these three fields. In discussing the details of each, two distinctions become of particular concern: first, a large part of the differing conceptions of "innate" seem to hinge on what is meant by "learning" and "acquiring," and therefore second, a fine line between UBL and Emergentism can be drawn, a relationship that is otherwise conflated in the literature. Because chapter two involves a brief account of the way in which connectionist simulations are often utilized to model or represent language acquisition, particularly from an Emergentist perspective, chapter three begins by examining this feature of Emergentism in more detail. Due to its explanatory power, ability to be effectively modeled, and the evidentiary support found in neuroscience, Emergentism would appear to be the most tenable position to maintain regarding language acquisition. This possibility seems further strengthened when we take into account the neuroscientific data often used to bolster anti-nativist claims. Nevertheless, reflecting on the overarching concern of the project, regarding what is really meant by "innate," it is shown that this attack on nativism might stand on shakier ground than was originally assumed. Finally, based on these considerations, a case is made for an intermediary position, a theory of "Minimally Innate Ideas.". (shrink)

We introduce a set of biologically and computationally motivated design choices for modeling the learning of language, or of other types of sequential, hierarchically structured experience and behavior, and describe an implemented system that conforms to these choices and is capable of unsupervised learning from raw natural-language corpora. Given a stream of linguistic input, our model incrementally learns a grammar that captures its statistical patterns, which can then be used to parse or generate new data. The grammar constructed in this (...) manner takes the form of a directed weighted graph, whose nodes are recursively (hierarchically) defined patterns over the elements of the input stream. We evaluated the model in seventeen experiments, grouped into five studies, which examined, respectively, (a) the generative ability of grammar learned from a corpus of natural language, (b) the characteristics of the learned representation, (c) sequence segmentation and chunking, (d) artificial grammar learning, and (e) certain types of structure dependence. The model's performance largely vindicates our design choices, suggesting that progress in modeling language acquisition can be made on a broad front—ranging from issues of generativity to the replication of human experimental findings—by bringing biological and computational considerations, as well as lessons from prior efforts, to bear on the modeling approach. (shrink)

There is an active debate within the field of phonology concerning the cognitive status of substantive phonetic factors such as ease of articulation and perceptual distinctiveness. A new framework is proposed in which substance acts as a bias, or prior, on phonological learning. Two experiments tested this framework with a method in which participants are first provided highly impoverished evidence of a new phonological pattern, and then tested on how they extend this pattern to novel contexts and novel sounds. Participants (...) were found to generalize velar palatalization (e.g., the change from [k] as in keep to [t⌢∫S] as in cheap) in a way that accords with linguistic typology, and that is predicted by a cognitive bias in favor of changes that relate perceptually similar sounds. Velar palatalization was extended from the mid front vowel context (i.e., before [e] as in cape) to the high front vowel context (i.e., before [i] as in keep), but not vice versa. The key explanatory notion of perceptual similarity is quantified with a psychological model of categorization, and the substantively biased framework is formalized as a conditional random field. Implications of these results for the debate on substance, theories of phonological generalization, and the formalization of similarity are discussed. (shrink)

Imitation has been understood in different ways: as a cognitive adaptation subtended by genetically specified cognitive mechanisms; as an aspect of domain general human cognition. The second option has been advanced by Cecilia Heyes who treats imitation as an instance of associative learning. Her argument is part of a deflationary treatment of the “mirror neuron” phenomenon. I agree with Heyes about mirror neurons but argue that Kim Sterelny has provided the tools to provide a better account of the nature and (...) role of human imitation. What we call imitative learning is an instance of social learning. It has little to do with empathy, emotional contagion, or mind reading. (shrink)

Our response takes advantage of the wide-ranging commentary to clarify some aspects of our original proposal and augment others. We argue against the generative critics of our coevolutionary program for the language sciences, defend the use of close-to-surface models as minimizing cross-linguistic data distortion, and stress the growing role of stochastic simulations in making generalized historical accounts testable. These methods lead the search for general principles away from idealized representations and towards selective processes. Putting cultural evolution central in understanding language (...) diversity makes learning fundamental in the cognition of language: increasingly powerful models of general learning, paired with channelled caregiver input, seem set to manage language acquisition without recourse to any innate “universal grammar.” Understanding why human language has no clear parallels in the animal world requires a cross-species perspective: crucial ingredients are vocal learning (for which there are clear non-primate parallels) and an intention-attributing cognitive infrastructure that provides a universal base for language evolution. We conclude by situating linguistic diversity within a broader trend towards understanding human cognition through the study of variation in, for example, human genetics, neurocognition, and psycholinguistic processing. (shrink)

Talk of linguistic universals has given cognitive scientists the impression that languages are all built to a common pattern. In fact, there are vanishingly few universals of language in the direct sense that all languages exhibit them. Instead, diversity can be found at almost every level of linguistic organization. This fundamentally changes the object of enquiry from a cognitive science perspective. This target article summarizes decades of cross-linguistic work by typologists and descriptive linguists, showing just how few and unprofound the (...) universal characteristics of language are, once we honestly confront the diversity offered to us by the world's 6,000 to 8,000 languages. After surveying the various uses of “universal,” we illustrate the ways languages vary radically in sound, meaning, and syntactic organization, and then we examine in more detail the core grammatical machinery of recursion, constituency, and grammatical relations. Although there are significant recurrent patterns in organization, these are better explained as stable engineering solutions satisfying multiple design constraints, reflecting both cultural-historical factors and the constraints of human cognition.Linguistic diversity then becomes the crucial datum for cognitive science: we are the only species with a communication system that is fundamentally variable at all levels. Recognizing the true extent of structural diversity in human language opens up exciting new research directions for cognitive scientists, offering thousands of different natural experiments given by different languages, with new opportunities for dialogue with biological paradigms concerned with change and diversity, and confronting us with the extraordinary plasticity of the highest human skills. (shrink)

Griffiths and Machery (2008) have argued that innateness is a folk notion that obstructs inquiry and has no place in contemporary science. They support their view by criticizing the canalization account of innateness (Ariew, 1999, 2006). In response, I argue that the criticisms they raise for the canalization account can be avoided by another recent account of innateness, the triggering account, which provides an analysis of the concept as it applies to cognitive capacities (Khalidi, 2002, 2007; Stich, 1975). I also (...) claim that they have not demonstrated that the folk notion of innateness is unsuitable for rehabilitation in a science of cognition. I conclude that they have not made the case that the notion of innateness ought to be eliminated from a scientific account of the mind. (shrink)

This paper attempts to articulate a dispositional account of innateness that applies to cognitive capacities. After criticizing an alternative account of innateness proposed by Cowie (1999) and Samuels (2002), the dispositional account of innateness is explicated and defended against a number of objections. The dispositional account states that an innate cognitive capacity (output) is one that has a tendency to be triggered as a result of impoverished environmental conditions (input). Hence, the challenge is to demonstrate how the input can be (...) compared to the output and shown to be relatively impoverished. I argue that there are robust methods of comparing input to output without measuring them quantitatively. (shrink)

Although for many years a sharp distinction has been made in language research between rules and words—with primary interest on rules—this distinction is now blurred in many theories. If anything, the focus of attention has shifted in recent years in favor of words. Results from many different areas of language research suggest that the lexicon is representationally rich, that it is the source of much productive behavior, and that lexically specific information plays a critical and early role in the interpretation (...) of grammatical structure. But how much information can or should be placed in the lexicon? This is the question I address here. I review a set of studies whose results indicate that event knowledge plays a significant role in early stages of sentence processing and structural analysis. This poses a conundrum for traditional views of the lexicon. Either the lexicon must be expanded to include factors that do not plausibly seem to belong there; or else virtually all information about word meaning is removed, leaving the lexicon impoverished. I suggest a third alternative, which provides a way to account for lexical knowledge without a mental lexicon. (shrink)

While rules and exemplars are usually viewed as opposites, this paper argues that they form end points of the same distribution. By representing both rules and exemplars as (partial) trees, we can take into account the fluid middle ground between the two extremes. This insight is the starting point for a new theory of language learning that is based on the following idea: If a language learner does not know which phrase‐structure trees should be assigned to initial sentences, s/he allows (...) (implicitly) for all possible trees and lets linguistic experience decide which is the “best” tree for each sentence. The best tree is obtained by maximizing “structural analogy” between a sentence and previous sentences, which is formalized by the most probable shortest combination of subtrees from all trees of previous sentences. Corpus‐based experiments with this model on the Penn Treebank and the Childes database indicate that it can learn both exemplar‐based and rule‐based aspects of language, ranging from phrasal verbs to auxiliary fronting. By having learned the syntactic structures of sentences, we have also learned the grammar implicit in these structures, which can in turn be used to produce new sentences. We show that our model mimicks children’s language development from item‐based constructions to abstract constructions, and that the model can simulate some of the errors made by children in producing complex questions. (shrink)

Recent research suggests that language evolution is a process of cultural change, in which linguistic structures are shaped through repeated cycles of learning and use by domain-general mechanisms. This paper draws out the implications of this viewpoint for understanding the problem of language acquisition, which is cast in a new, and much more tractable, form. In essence, the child faces a problem of induction, where the objective is to coordinate with others (C-induction), rather than to model the structure of the (...) natural world (N-induction). We argue that, of the two, C-induction is dramatically easier. More broadly, we argue that understanding the acquisition of any cultural form, whether linguistic or otherwise, during development, requires considering the corresponding question of how that cultural form arose through processes of cultural evolution. This perspective helps resolve the “logical” problem of language acquisition and has far-reaching implications for evolutionary psychology. (shrink)

This article examines and rejects the claim that 'innateness is canalization'. Waddington's concept of canalization is distinguished from the narrower concept of environmental canalization with which it is often confused. Evidence is presented that the concept of environmental canalization is not an accurate analysis of the existing concept of innateness. The strategy of 'biologicizing the mind' by treating psychological or behavioral traits as if they were environmentally canalized physiological traits is criticized using data from developmental psychobiology. It is concluded that (...) identifying innateness with environmental canalization can only result in adding unhelpful associations from 'folkbiology' to the relatively precise idea of canalization. (shrink)

According to, when forming complex yes/no questions, children do not make errors such as Is the boy who smoking is crazy? because they have innate knowledge of structure dependence and so will not move the auxiliary from the relative clause. However, simple recurrent networks are also able to avoid such errors, on the basis of surface distributional properties of the input (; ). Two new elicited production studies revealed that (a) children occasionally produce structure‐dependence errors and (b) the pattern of (...) children's auxiliary‐doubling errors (Is the boy who is smoking is crazy?) suggests a sensitivity to surface co‐occurrence patterns in the input. This article concludes that current data do not provide any support for the claim that structure dependence is an innate constraint, and that it is possible that children form a structure‐dependent grammar on the basis of exposure to input that exhibits this property. (shrink)

A new theoretical approach to language has emerged in the past 10–15 years that allows linguistic observations about form–meaning pairings, known as ‘construc- tions’, to be stated directly. Constructionist approaches aim to account for the full range of facts about language, without assuming that a particular subset of the data is part of a privileged ‘core’. Researchers in this field argue that unusual constructions shed light on more general issues, and can illuminate what is required for a complete account of (...) language. (shrink)

According to classical arguments, language learning is both facilitated and constrained by cognitive biases. These biases are reflected in linguistic typology—the distribution of linguistic patterns across the world's languages—and can be probed with artificial grammar experiments on child and adult learners. Beginning with a widely successful approach to typology (Optimality Theory), and adapting techniques from computational approaches to statistical learning, we develop a Bayesian model of cognitive biases and show that it accounts for the detailed pattern of results of artificial (...) grammar experiments on noun-phrase word order (Culbertson, Smolensky, & Legendre, 2012). Our proposal has several novel properties that distinguish it from prior work in the domains of linguistic theory, computational cognitive science, and machine learning. This study illustrates how ideas from these domains can be synthesized into a model of language learning in which biases range in strength from hard (absolute) to soft (statistical), and in which language-specific and domain-general biases combine to account for data from the macro-level scale of typological distribution to the micro-level scale of learning by individuals. (shrink)

A central goal of modern generative grammar has been to discover invariant properties of human languages that reflect “the innate schematism of mind that is applied to the data of experience” and that “might reasonably be attributed to the organism itself as its contribution to the task of the acquisition of knowledge” (Chomsky, 1971). Candidates for such invariances include the structure dependence of grammatical rules, and in particular, certain constraints on question formation. Various “poverty of stimulus” (POS) arguments suggest that (...) these invariances reflect an innate human endowment, as opposed to common experience: Such experience warrants selection of the grammars acquired only if humans assume, a priori, that selectable grammars respect substantive constraints. Recently, several researchers have tried to rebut these POS arguments. In response, we illustrate why POS arguments remain an important source of support for appeal to a priori structure-dependent constraints on the grammars that humans naturally acquire. (shrink)

Our target article argued that a genetically specified Universal Grammar (UG), capturing arbitrary properties of languages, is not tenable on evolutionary grounds, and that the close fit between language and language learners arises because language is shaped by the brain, rather than the reverse. Few commentaries defend a genetically specified UG. Some commentators argue that we underestimate the importance of processes of cultural transmission; some propose additional cognitive and brain mechanisms that may constrain language and perhaps differentiate humans from nonhuman (...) primates; and others argue that we overstate or understate the case against co-evolution of language genes. In engaging with these issues, we suggest that a new synthesis concerning the relationship between brains, genes, and language may be emerging. (shrink)

Christiansen & Chater (C&C) argue persuasively that Universal Grammar (UG) could not have arisen through evolutionary processes. I provide additional suggestions to strengthen the argument against UG evolution. Further, I suggest that C&C's solution to the logical problem of language evolution faces several problems. Widening the focus to mechanisms of general cognition and inclusion of animal communication research might overcome these problems.

I object to eliminativism about innateness and André Ariew’s identification of innateness with canalization, and I propose a new treatment of innateness. I first argue that the concept of innateness is serving a valuable function in a diverse set of research contexts, and in these contexts, claims about innateness are best understood as claims about the insensitivity of the appearance of a trait to certain variations in the environment. I then argue that innateness claims, like claims about canalization, should be (...) explicitly relativized to the specific range of environmental variations of interest to the scientist. My account characterizes an important way in which scientists are employing the concept and offers a way for scientists to carry on using the concept in their research while minimizing confusion and miscommunication. There is a fruitful research program, I claim, in which scientists employ the concept of innateness to help distinguish environmental factors of interest that have a causal influence on the appearance of a trait from those that do not. (shrink)

The poverty of stimulus argument is one of the most controversial arguments in the study of language acquisition. Here we follow previous approaches challenging the assumption of impoverished primary linguistic data, focusing on the specific problem of auxiliary (AUX) fronting in complex polar interrogatives. We develop a series of corpus analyses of child‐directed speech showing that there is indirect statistical information useful for correct auxiliary fronting in polar interrogatives and that such information is sufficient for distinguishing between grammatical and ungrammatical (...) generalizations, even in the absence of direct evidence. We further show that there are simple learning devices, such as neural networks, capable of exploiting such statistical cues, producing a bias toward correct AUX questions when compared to their ungrammatical counterparts. The results suggest that the basic assumptions of the poverty of stimulus argument may need to be reappraised. (shrink)

Natural language is full of patterns that appear to fit with general linguistic rules but are ungrammatical. There has been much debate over how children acquire these “linguistic restrictions,” and whether innate language knowledge is needed. Recently, it has been shown that restrictions in language can be learned asymptotically via probabilistic inference using the minimum description length (MDL) principle. Here, we extend the MDL approach to give a simple and practical methodology for estimating how much linguistic data are required to (...) learn a particular linguistic restriction. Our method provides a new research tool, allowing arguments about natural language learnability to be made explicit and quantified for the first time. We apply this method to a range of classic puzzles in language acquisition. We find some linguistic rules appear easily statistically learnable from language experience only, whereas others appear to require additional learning mechanisms (e.g., additional cues or innate constraints). (shrink)

In the present study, using event-related functional magnetic resonance imaging, we investigated a group of participants on a grammaticality classification task after they had been exposed to well-formed consonant strings generated from an artificial regular grammar. We used an implicit acquisition paradigm in which the participants were exposed to positive examples. The objective of this studywas to investigate whether brain regions related to language processing overlap with the brain regions activated by the grammaticality classification task used in the present study. (...) Recent meta-analyses of functional neuroimaging studies indicate that syntactic processing is related to the left inferior frontal gyrus (Brodmann's areas 44 and 45) or Broca's region. In the present study, we observed that artificial grammaticality violations activated Broca's region in all participants. This observation lends some support to the suggestions that artificial grammar learning represents a model for investigating aspects of language learning in infants [TICS 4 (2000) 178] and adults [Proceedings of the National Academy of Sciences of United States of America 99 (2002) 529]. (shrink)

We introduce a set of biologically and computationally motivated design choices for modeling the learning of language, or of other types of sequential, hierarchically structured experience and behavior, and describe an implemented system that conforms to these choices and is capable of unsupervised learning from raw natural‐language corpora. Given a stream of linguistic input, our model incrementally learns a grammar that captures its statistical patterns, which can then be used to parse or generate new data. The grammar constructed in this (...) manner takes the form of a directed weighted graph, whose nodes are recursively (hierarchically) defined patterns over the elements of the input stream. We evaluated the model in seventeen experiments, grouped into five studies, which examined, respectively, (a) the generative ability of grammar learned from a corpus of natural language, (b) the characteristics of the learned representation, (c) sequence segmentation and chunking, (d) artificial grammar learning, and (e) certain types of structure dependence. The model's performance largely vindicates our design choices, suggesting that progress in modeling language acquisition can be made on a broad front—ranging from issues of generativity to the replication of human experimental findings—by bringing biological and computational considerations, as well as lessons from prior efforts, to bear on the modeling approach. (shrink)

Wittgenstein's notorious sample of a ‘complete primitive language’ is often thought to be closer in kind to animal forms of communication than human language. Indeed, it has been criticised on precisely these grounds. But such debates make little sense if we take seriously Wittgenstein's idea that language is a family resemblance concept. So, rather than argue that the builders’ game ‘really is a language’, I propose to turn the debate on its head and welcome the comparison. By changing our perspective (...) in this way, I suggest that we can see that the learning of language is crucially dependent on forms of communication that are animal in nature. I then discuss how these lessons might shed light on empirical research into both the ontogenetic and phylogenetic origins of linguistic communication. (shrink)

Using an asymptotic characterization of probabilistic finite state languages over a one-letter alphabet we construct a probabilistic language with regular support that cannot be generated by probabilistic CFGs. Since all probability values used in the example are rational, our work is immune to the criticism leveled by Suppes (Synthese 22:95–116, 1970 ) against the work of Ellis ( 1969 ) who first constructed probabilistic FSLs that admit no probabilistic FSGs. Some implications for probabilistic language modeling by HMMs are discussed.

This paper focuses on the linguistic evidence base provided by proponents of conceptualism (e.g., Chomsky) and rational realism (e.g., Katz) and challenges some of the arguments alleging that the evidence allowed by conceptualists is superior to that of rational realists. Three points support this challenge. First, neither conceptualists nor realists are in a position to offer direct evidence. This challenges the conceptualists’ claim that their evidence is inherently superior. Differences between the kinds of available indirect evidence will be discussed. Second, (...) at least some of the empirical evidence provided by the conceptualist is flawed. It is not obtained independently of theoretical commitments, alternative interpretations have not been ruled out, and some of the thought experiments intended to extend the evidence base are conceptually flawed. Third, the widely held assumption that rational realism disallows empirical evidence relevant to linguistics is dubious. It will be shown that the limitation imposed by rational realism concerns strictly formal linguistics. The rationalist realist has no reason to impose any restriction on the evidence relevant to psycholinguistics. I conclude that it is a mistake to dismiss realism based on the assumption that it imposes undue restrictions on evidence that is relevant to linguistics. (shrink)