Connectionist semantic systematicity

Fodor and Pylyshyn [Fodor, J. A., & Pylyshyn, Z. W. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28, 3–71] argue that connectionist models are not able to display systematicity other than by implementing a classical symbol system. This claim entails that connectionism cannot compete with the classical approach as an alternative architectural framework for human cognition. We present a connectionist model of sentence comprehension that does not implement a symbol system yet behaves systematically. It consists in a recurrent neural network that maps sentences describing situations in a microworld, onto representations of these situations. After being trained on particular sentence–situation pairs, the model can comprehend new sentences, even if these describe new situations. We argue that this systematicity arises robustly and in a psychologically plausible manner because it depends on structure inherent in the world.     

Three-concept monte: Explanation,implementation and systematicity

Fodor and Pylyshyn (1988), Fodor and McLaughlin (1990) and McLaughlin (1993) challenge connectionists to explain systematicity without simply implementing a classical architecture. In this paper I argue that what makes the challenge difficult for connectionists to meet has less to do with what is to be explained than with what is to count as an explanation. Fodor et al. are prepared to admit as explanatory, accounts of a sort that only classical models can provide. If connectionists are to meet the challenge, they are going to have to insist on the propriety of changing what counts as an explanation of systematicity. Once that is done, there would seem to be as yet no reason to suppose that connectionists are unable to explain systematicity.Shorter versions of this paper were presented at the 1993 annual meetings of both the Southern Society for Philosophy and Psychology and the Society for Philosophy and Psychology. The paper has benefited from the comments and criticisms of Frances Egan, David Chalmers, Terry Horgan, Brian McLaughlin, Tim van Gelder and Ted Warfield.     

Connectionism and the problem of systematicity: why Smolensky's solution doesn't work

In two recent papers, Paul Smolensky responds to a challenge Jerry Fodor and Zenon Pylyshyn posed for connectionist theories of cognition: to explain the existence of systematic relations among cognitive capacities without assuming that mental processes are causally sensitive to the constituent structure of mental representations. Smolensky thinks connectionists can explain systematicity if they avail themselves of "distributed" mental representation. In facts, Smolensky offers two accounts of distributed mental representation, corresponding to his notions of "weak" and "strong" compositional structure. We argue that weak compositional structure is irrelevant to the systematicity problem and of dubious internal coherence. We then argue that strong compositional (tensor product) representations fail to explain systematicity because they fail to exhibit the sort of constituents that can provide domains for structure sensitive mental processes.     

Brain regions as difference-makers

Contrastive neuroimaging is often taken to provide evidence about the localization of cognitive functions. After canvassing some problems with this approach, I offer an alternative: neuroimaging gives evidence about regions of the brain that bear difference-making relationships to psychological processes of interest. I distinguish between the specificity and what I call the systematicity of a difference-making relationship, and I show how at least some neuroimaging experiments can give evidence for systematic difference-making.     

Cognition without classical architecture

Fodor and Pylyshyn (1988) argue that any successful model of cognition must use classical architecture; it must depend upon rule-based processing sensitive to constituent structure. This claim is central to their defense of classical AI against the recent enthusiasm for connectionism. Connectionist nets, they contend, may serve as theories of the implementation of cognition, but never as proper theories of psychology. Connectionist models are doomed to describing the brain at the wrong level, leaving the classical view to account for the mind.This paper considers whether recent results in connectionist research weigh against Fodor and Pylyshyn's thesis. The investigation will force us to develop criteria for determining exactly when a net is capable of systematic processing. Fodor and Pylyshyn clearly intend their thesis to affect the course of research in psychology. I will argue that when systematicity is defined in a way that makes the thesis relevant in this way, the thesis is challenged by recent progress in connectionism.     

Talking our way to systematicity

Do we think in a language-like format? Taking the marker of language-like formats to be the property of unconstrained systematicity, this paper considers the following master argument for the claim that we do: (1) language is unconstrainedly systematic, (2) if language is unconstrainedly systematic then so is thought, (3) so thought is unconstrainedly systematic. It is easy to feel that there is something right about this argument, that there will be some way of filling in its details that will vindicate the idea that our thought must be unconstrainedly systematic given that the language in which we express it is. Clearly, however, the second premise needs support—we need a principled reason for moving from the unconstrained systematicity of language to the unconstrained systematicity of thought. This paper gives three passes at formulating such a principle. This turns out to be much harder than it might seem. We should, I conclude, resist falling too easily for the lure of this master argument for the language-like format of thought.

     

Articulating an Uncompromising Forgiveness

I first pose a challenge which, it seems to me, any philosophical account of forgiveness must meet: the account must be articulate and it must allow for forgiveness that is uncompromising. I then examine an account of forgiveness (proposed by David Novitz in the June 1998 issue of this journal) which appears to meet this challenge. Upon closer examination we discover that this account actually fails to meet the challenge—but it fails in very instructive ways. the account takes two missteps which seem to be taken by almost everyone discussing forgiveness. At the end, I sketch an alternative account of forgiveness, one that I think meets the challenge and avoids the missteps.     

Natural deduction in connectionist systems

The relation between logic and thought has long been controversial, but has recently influenced theorizing about the nature of mental processes in cognitive science. One prominent tradition argues that to explain the systematicity of thought we must posit syntactically structured representations inside the cognitive system which can be operated upon by structure sensitive rules similar to those employed in systems of natural deduction. I have argued elsewhere that the systematicity of human thought might better be explained as resulting from the fact that we have learned natural languages which are themselves syntactically structured. According to this view, symbols of natural language are external to the cognitive processing system and what the cognitive system must learn to do is produce and comprehend such symbols. In this paper I pursue that idea by arguing that ability in natural deduction itself may rely on pattern recognition abilities that enable us to operate on external symbols rather than encodings of rules that might be applied to internal representations. To support this suggestion, I present a series of experiments with connectionist networks that have been trained to construct simple natural deductions in sentential logic. These networks not only succeed in reconstructing the derivations on which they have been trained, but in constructing new derivations that are only similar to the ones on which they have been trained.     

Connectionism and the Philosophical Foundations of Cognitive Science

This is an overview of recent philosophical discussion about connectionism and the foundations of cognitive science. Connectionist modeling in cognitive science is described. Three broad conceptions of the mind are characterized, and their comparative strengths and weaknesses are discussed: (1) the classical computation conception in cognitive science; (2) a popular foundational interpretation of connectionism that John Tienson and I call "non-sentential computationalism"; and (3) an alternative interpretation of connectionism we call "dynamical cognition." Also discussed are two recent philosophical attempts to enlist connectionism in defense of eliminativism about folk psychology.     

McKinsey Paradoxes,Radical Scepticism,and the Transmission of Knowledge across Known Entailments

A great deal of discussion in the recent literature has been devoted to the so-called `McKinsey' paradox which purports to show that semantic externalism is incompatible with the sort of authoritative knowledge that we take ourselves to have of our own thought contents. In this paper I examine one influential epistemological response to this paradox which is due to Crispin Wright and Martin Davies. I argue that it fails to meet the challenge posed by McKinsey but that, if it is set within an externalist epistemology, it may have application to a related paradox that concerns the problem of radical scepticism.     

Can Compositionality Solve the Thought-or-Language Problem?

Abstract

Jerry Fodor has claimed to have a solution to the traditional problem of what comes first, thought or language. Compositionality, he says, will give us the answer, for at least one must be compositional, and if only one of them is, that is the one that has underived semantic content. He argues that natural languages are not compositional, and therefore that the content of language is derived from the content of thought. I will argue that the idea that language is not compositional conflicts with his productivity and systematicity arguments for the existence of a language of thought. I will also show that Fodor’s solution to the problem fails, as his main argument is circular. Finally, I suggest that Fodor’s argument against the compositionality of language is not decisive, and that we can still attribute at least some degree of compositionality to language.     

Connectionist representation,semantic compositionality,and the instability of concept structure

Summary In the present paper connectionist approaches to the problem of internal representation and the nature of concepts are discussed. In the first part the concept of representation that underlies connectionist modeling is made explicit. It is argued that the connectionist view of representation relies on a correlational theory of semantic content- i.e., the covariation between internal and external states is taken as the basis for ascribing meaning to internal states. The problems and virtues of such a correlational approach to internal representation are addressed. The second part of the paper is concerned with whether connectionism is capable of accounting for the apparent productivity and systematicity of language and thought. There is an evaluation of the recent arguments of Fodor and Pylyshyn, who claim that systematicity can only be explained if one conceives of mental representations as structured symbols composed of context-free constituents. There is a review of empirical evidence that strongly suggests that concepts are not fixed memory structures and that the meaning of constituent symbols varies, depending on the context in which they are embedded. On the basis of this review it is concluded that the meaning of a complex expression is not computed from the context-free meanings of the constituents, and that strong compositionality, as endorsed by Fodor and Pylyshyn (1988), seems implausible as a process theory for the comprehension of complex concepts. Instead, the hypothesis is endorsed that constraint satisfaction in distributed connectionist networks may allow for an alternative account of weak compositionality compatible with the context sensitivity of meaning. In the final section, it is argued that neither mere implementation of a language of thought in connectionist networks nor radical elimination of symbol systems seems to be a fruitful research strategy, but that it might be more useful to discuss how connectionist systems can develop the capacity to use external symbol systems like language or logic without instantiating symbol systems themselves.     

Directions in Connectionist Research: Tractable Computations Without Syntactically Structured Representations

Should connectionists abandon the quest for tractably computable cognitive transition functions while retaining syntactically structured mental representations? We argue, in opposition to Horgan, that it should not. We argue that the case against tractably computable functions, based upon the claimed isotropic and Quinean character of cognition, fails since cognition is not as isotropic and Quinean as Fodor and Horgan contend. Moreover, we illustrate how current research in both connectionism and cognitive neuroscience suggests that tractability can be preserved through division of overall computations into modular sub-processes, each of which is tractable. As to syntactically structured representations, we argue that they are unneeded for most cognitive tasks organisms confront, and that when they are needed, they may be provided by external representational media such as natural language. Moreover, we note that increasingly cognitive linguistics has become the ally of connectionism and that the research program of cognitive linguistics suggests that abilities to use natural languages may be developed without requiring syntactically structured mental representations to exist prior to natural language.     

The Value of Symbolic Computation

Standard generative linguistic theory, which uses discrete symbolic models of cognition, has some strengths and weaknesses. It is strong on providing a network of outposts that make scientific travel in the jungles of natural language feasible. It is weak in that it currently depends on the elaborate and unformalized use of intuition to develop critical supporting assumptions about each data point. In this regard, it is not in a position to characterize natural language systems in the lawful terms that ecological psychologists strive for. Connectionist learning models offer some help: They define lawful relations between linguistic environments and language systems. But our understanding of them is currently weak, especially when it comes to natural language syntax. Fortunately, symbolic linguistic analysis can help connectionism if the two meet via dynamical systems theory. I discuss a case in point: Insights from linguistic explorations of natural language syntax appear to have identified information structures that are particularly relevant to understanding ecologically appealing but analytically mysterious connectionist learning models.     

How Otto did not extend his mind,but might have: Dynamic systems theory and social-cultural group selection

Proponents of cognitive Situationism argue that the human mind is embodied, embedded in both natural and social-cultural environments and extended creating both extended and distributed cognition. Anti-situationists reject all or some of these claims. I argue that four major objections to extended cognition: (1) the mark of the cognitive, (2) the function-identity fallacy, (3) cognitive bloat, and (4) scientific irrelevance lose much of their sting in the case of distributed cognition, the extension of cognitive agency to a group of cognitive agents, such as a scientific research team. However, I claim that a crucial fifth challenge, that advocates of the extended mind commit the causal-constitution fallacy, has yet to be satisfactorily addressed. I focus on Spyridon Palermos’ use of dynamic systems theory to refute this charge and I argue that his appeal to dynamic systems theory as a way of understanding system-constitution fails. Instead, I suggest a social-cultural group selection hypothesis for understanding system-constitution. But, I leave it for another day to elaborate that hypothesis’ empirical plausibility.     

Meta‐normative Realism,Evolution, and Our Reasons to Survive

In this article, I articulate and respond to an epistemological challenge to meta‐normative realism. The challenge has it that, if realism about the normative is correct, and if evolutionary forces have significantly influenced our normative judgments, then it would be a remarkable coincidence if the content of the normative facts and our normative judgments were aligned. I criticize David Enoch's recent attempt to meet this challenge, but provide an alternative response that is structurally similar. I argue that if realism is correct, then it would be remarkable if the content of our normative judgments and the normative facts were not significantly aligned.     

Externalism,Slow Switching and Privileged Self‐Knowledge1

Recent discussions of externalism about mental content have been dominated by the question whether it undermines the intuitively plausible idea that we have knowledge of the contents of our thoughts. In this article I focus on one main line of reasoning (the so‐called ‘slow switching argument’) for the thesis that externalism and self‐knowledge are incompatible. After criticizing a number of influential responses to the argument, I set out to explain why it fails. It will be claimed that the argument trades on an ambiguity, and that only by incorporating certain controversial assumptions does it stand a chance of establishing its conclusion. Finally, drawing on an analogy with Benacerraf's challenge to Platonism, I shall offer some reasons as to why the slow switching argument fails to reveal the real source of tension between externalism and privileged self‐knowledge.     

On Aesthetic Judgement and our Relation to Nature: Kant's Concept of Purposiveness

I offer a critical reconstruction of Kant's thesis that aesthetic judgement is founded on the principle of the purposiveness of nature. This has been taken as equivalent to the claim that aesthetics is directly linked to the systematicity of nature in its empirical laws. I take issue both with Henry Allison, who seeks to marginalize this claim, and with Avner Baz, who highlights it in order to argue that Kant's aesthetics are merely instrumental for his epistemology. My solution is that aesthetic judgement operates as an exemplary presentation of our general ability to schematise an intuition with a concept at the empirical level. I suggest that this counts as an empirical schematism. Although aesthetic judgement is not based on empirical systematicity, it can nevertheless offer indirect support for the latter in so far as it is a particular revelation of purposiveness in general.     

Mental Maps1

It’s often hypothesized that the structure of mental representation is map‐like rather than language‐like. The possibility arises as a counterexample to the argument from the best explanation of productivity and systematicity to the language of thought hypothesis—the hypothesis that mental structure is compositional and recursive. In this paper, I argue that the analogy with maps does not undermine the argument, because maps and language have the same kind of compositional and recursive structure.