We describe a deep generative model in which the lowest layer represents the word-count vector of a document and the top layer represents a learned binary code for that document. The top two layers of the generative model form an undirected associative memory and the remaining layers form a belief net with directed, top-down connections. We present efficient learning and inference procedures for this type of generative model and show that it allows more accurate and much faster retrieval than latent (...) semantic analysis. By using our method as a filter for a much slower method called TF-IDF we achieve higher accuracy than TF-IDF alone and save several orders of magnitude in retrieval time. By using short binary codes as addresses, we can perform retrieval on very large document sets in a time that is independent of the size of the document set using only one word of memory to describe each document. (shrink)

Disjunctivism, as a theory of visual experience, claims that the mental states involved in a “good case” experience of veridical perception and a “bad case” experience of hallucination differ, even in those cases in which the two experiences are indistinguishable for their subject. Consider the veridical perception of a bar stool and an indistinguishable hallucination; both of these experiences might be classed together as experiences (as) of a bar stool or experiences of seeming to see a bar stool. This might (...) lead us to think that the experiences we undergo in the two cases must be of the same kind, the difference being that the former, but not the latter, is connected to the world in the right kind of way. Such a conjecture has been called a “highest common factor” or “common kind” assumption. At heart, disjunctivism consists in the rejection of this assumption. According the disjunctivist, veridical experiences and hallucinations do not share a common component. There are a host of interesting questions surrounding disjunctivism including: What is involved in the claims that good case and bad case experiences differ? Why might one might want to be a disjunctivist? What kinds of claims can the disjunctivist make about hallucination and illusion? These questions, and problems for the thesis, will be discussed as we proceed. (shrink)

In company with its sister volume, Arts-Based Research Across Textual Media in Education explores arts-based approaches to research across media, including film and comics-related material, from a variety of geographic locations and across a range of sub-disciplines within the field of education. This first volume takes a textual focus, capturing process, poetic, and dramaturgical approaches. The authors aim to highlight some of the approaches that are not always centered in arts-based research. The contributors represent a variety of arts-based practices and (...) methods, and they weave this marrying of artistic and scientific expertise and experience into the fabric of the chapters themselves. Authors from international contexts speak to the importance of utilizing artistic approaches for research processes. From multimodal field notes to poetic forms to the dramaturgical, chapters in this book represent steps forward in educational inquiry to bringing together both the creative and credible. The book includes multiple images and rich descriptions shared from the field. This first volume covers amongst other topics: co-created narratives; creative fiction in research; analytic portraits; dramatic representation, and critical poetic inquiry. It would be suitable for graduate students and scholars interested in qualitative inquiry and arts-based methods, in Education and the social sciences. (shrink)

am going to discuss some issues inspired by a well-known paper ofKeith Donnellan, "Reference and Definite Descriptions,”2 but the interest—to me—of the contrast mentioned in my title goes beyond Donnellan's paper: I think it is of considerable constructive as well as critical importance to the philosophy oflanguage. These applications, however, and even everything I might want to say relative to Donnellan’s paper, cannot be discussed in full here because of problems of length. Moreover, although I have a considerable interest in (...) the substantive issues raised by Donnellan’s paper, and by related literature, my own conclusions will be methodological, not substantive. I can put the matter this way: Donnellan’s paper claims to give decisive objections both to Russell’s theory of definite descriptions (taken as a theory about English) and to Strawson’s. My concem is not primarily with the question; is Donnellan right, or is Russell (or Strawson)? Rather, it is with the question: do the considerations in Donneilarfs paper refute Russell’s theory (or Strawson’s)? For definiteness, I will concentrate on Donnellan versus Russell, leaving Strawson aside. And about this issue I will draw a definite conclusion, one which I think will illuminate a few methodological maxims about language. Namely, I will conclude that the considerations in Donnellan’s paper, by themselves, do not refute Russell’s theory. Any conclusions about Russell’s views per se, or Donnellan’s, must be tentative, IfI were to be asked for a tentative stab about Russell, I would say that although his theory does a far better job of handling ordinary discourse than many have thought, and although many popular arguments against it are inconclusive, probably it ultimately fails. The considerations I have in mind have to do with the existence of “improper” definite descriptions, such as “the table," where uniquely specifying conditions are not contained in the description itself.. (shrink)

The central aim of the inquiry begun in this text is to reach a semantic characterisationof philosophical discourse, that is, to describe the »language«, or the code, ofphilosophy. This inquiry contains an examination of the views on the nature andpurpose of philosophy held by Immanuel Kant and Ludwig Wittgenstein, but manyother philosophers, semioticians, linguists and literary theorists are brought into thediscussion.In the first part of the text, the view is expressed that, with regard to the peculiarphenomena that characterize philosophy (...) , a theory of philosophy itself is needed, but such that would notitself be caught in the same kind of discourse. Then some methodological restrictionsare introduced: mainly, that the »philosophy« to be dealt with is the classicalcontinental philosophy, which is percieved as a body of texts. The aim of the inquiryis then formulated as the description of the code by which these texts are organized;the method of the inquiry is specified as a deductive-hypothetical one. In the end,an outline of a semantic theory is offered, as a basis for this description.The second part of the text starts the examination of Kant’s and Wittgenstein’sphilosophical views and their views on philosophy. Kant’s theory of knowledge ispresented as one which asserts the primary role of subjective a priori forms whichshape what we know as »reality«, and is therefore opposed to Wittgenstein’s picturetheory of language, where language simply mirrors what is given. The conclusion isreached, however, that both Kant and Wittgenstein propose a parallel dichotomybetween that which can be known/said, and the thing-in-itself or the mystical, andwith regard to this they both claim that the aims of philosophy as seen traditionallymust be reconcieved to a great extent.The inquiry is continued in an another article. (shrink)

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 chapter defends the property dualism argument. The term “semantic premise” mentioned is used to refers to an assumption identified by Brian Loar that antiphysicalist arguments, such as the property dualism argument, tacitly assume that a statement of property identity that links conceptually independent concepts is true only if at least one concept picks out the property it refers to by connoting a contingent property of that property. It is argued that, the property that does the work in explaining (...) the possibility of a posteriori identities need not be a first-order property of the referent in question. On his view, the property dualism argument requires only a weaker semantic premise, which allows that the property in question be a higher order property. A refined version of the property dualism argument is formulated, which uses the weaker premise, and defends the argument against various objections. (shrink)

Hans van Ditmarsch and Barteld Kooi (2008). Semantic results for ontic and epistemic change. In: G. Bonanno, W. van der Hoek and M. Wooldridge (editors). Logic and the Foundations of Game and Decision Theory (LOFT 7). Texts in Logic and Games 3, pp. 87-117, Amsterdam University Press, Amsterdam.

This paper argues that several leading theories of subjunctive conditionals are incompatible with ordinary intuitions about what credences we ought to have in subjunctive conditionals. In short, our theory of subjunctives should intuitively display semantic humility, i.e. our semantic theory should deliver the truth conditions of sentences without pronouncing on whether those conditions actually obtain. In addition to describing intuitions about subjunctive conditionals, I argue that we can derive these ordinary intuitions from justified premises, and I answer a (...) possible worry for my derivation by refuting a subjunctive triviality result modeled on (Lewis 1976). I conclude that the debate over the correct theory of subjunctive conditionals requires settling meta-philosophical questions about the relative value of various virtues of first-order theories of subjunctive conditionals. (shrink)

In general, stimuli that are familiar and recognizable have an advantage of predominance during binocular rivalry. Recent research has demonstrated that familiar and recognizable stimuli such as upright faces and words in a native language could break interocular suppression faster than their matched controls. In this study, a visible word prime was presented binocularly then replaced by a high-contrast dynamic noise pattern presented to one eye and either a semantically related or unrelated word was introduced to the other eye. We (...) measured how long it took for target words to break from suppression. To investigate word-parts priming, a second experiment also included word pairs that had overlapping subword fragments. Results from both experiments consistently show that semantically related words and words that shared subword fragments were faster to gain dominance compared to unrelated words, suggesting that words, even when interocularly suppressed and invisible, can benefit from semantic and subword priming. (shrink)

The semantic view of computation is the claim that semantic properties play an essential role in the individuation of physical computing systems such as laptops and brains. The main argument for the semantic view rests on the fact that some physical systems simultaneously implement different automata at the same time, in the same space, and even in the very same physical properties. Recently, several authors have challenged this argument. They accept the premise of simultaneous implementation but reject (...) the semantic conclusion. In this paper, I aim to explicate the semantic view and to address these objections. I first characterize the semantic view and distinguish it from other, closely related views. Then, I contend that the master argument for the semantic view survives the counter-arguments against it. One counter-argument is that computational individuation is not forced to choose between the implemented automata but rather always picks out a more basic computational structure. My response is that this move might undermine the notion of computational equivalence. Another counter-argument is that while computational individuation is forced to rely on extrinsic features, these features need not be semantic. My reply is that the semantic view better accounts for these extrinsic features than the proposed non-semantic alternatives. (shrink)

Lewis signalling games are often used to explain how it is possible for simple agents to develop systems of conventional semantic meaning. In these games, all players obtain identical payoffs in every outcome. This is an unrealistic payoff structure, but it is often employed because it is thought that semantic meaning will not emerge if interests conflict. Here it is shown that not only is conventional meaning possible when interests conflict, but it is the most likely outcome in (...) a finite population model of learning known as the Moran process. On the basis of this result it is suggested that evolutionary game theory’s standard models may yield results that are systematically distorted for a class of signalling games that have the abstract structure of social dilemmas. 1 Introduction2 The Seduction Game3 Imitation Dynamics with Small Mutations4 Dynamics with Unverifiable Message5 Analysis of a Related Three-Strategy Game6 Conclusion. (shrink)

There has frequently been taken to be a tension, if not an incompatibility, between "externalist" theories of content (which allow the make-up of one's physical environment and the linguistic usage of one's community to contribute to the contents of one's thoughts and utterances) and the "methodologically individualist" intuition that whatever contributes to the content of one's thoughts and utterances must ultimately be grounded in facts about one's own attitudes and behavior. In this dissertation I argue that one can underwrite such (...) externalist theories within a methodologically individualistic framework by understanding semantic norms in terms of the need to reach, for each of one's terms, a type of "equilibrium." Each speaker's commitment to making her _own_ beliefs and applications consistent allows one to incorporate these 'external' factors into the contents of their thoughts and utterances in a way that remains methodologically individualistic. Methodologically individualistic accounts are typically taken to be unable to incorporate 'external' factors such as the world's physical make-up or communal usage because of arguments suggesting that the individual's own beliefs and usage underdetermine or even misidentify what, according to externalist accounts, they mean by their terms. These arguments, however, only seem plausible if one presupposes a comparatively impoverished conception of the individual's beliefs. The beliefs a speaker associates with a given term extend far beyond the handful of sentences they would produce if asked to list such beliefs. In particular, speakers have an implicit, but rich, understanding of their language, their world, and the relation between them. Speakers typically understand languages as shared temporally extended practices about which they can be, both individually and collectively, mistaken. Once this conception of language is taken into account, the ascriptions which purportedly forced 'non-individualistic' conceptions of content upon us (particularly ascriptions which seemed to tie what we meant to social use rather than our own beliefs) turn out to be ultimately grounded in the individual's own beliefs. Indeed, our self-conception does much more than merely underwrite 'non-individualistic' ascriptions.. (shrink)

Subjects classified visible 2-digit numbers as larger or smaller than 55. Target numbers were preceded by masked 2-digit primes that were either congruent (same relation to 55) or incongruent. Experiments 1 and 2 showed prime congruency effects for stimuli never included in the set of classified visible targets, indicating subliminal priming based on long-term semantic memory. Experiments 2 and 3 went further to demonstrate paradoxical unconscious priming effects resulting from task context. For example, after repeated practice classifying 73 as (...) larger than 55, the novel masked prime 37 paradoxically facilitated the “larger” response. In these experiments task context could induce subjects to unconsciously process only the leftmost masked prime digit, only the rightmost digit, or both independently. Across 3 experiments, subliminal priming was governed by both task context and long-term semantic memory. (shrink)

Aggregating snippets from the semantic memories of many individuals may not yield a good map of an individual’s semantic memory. The authors analyze the structure of semantic networks that they sampled from individuals through a new snowball sampling paradigm during approximately 6 weeks of 1-hr daily sessions. The semantic networks of individuals have a small-world structure with short distances between words and high clustering. The distribution of links follows a power law truncated by an exponential cutoff, (...) meaning that most words are poorly connected and a minority of words has a high, although bounded, number of connections. Existing aggregate networks mirror the individual link distributions, and so they are not scale-free, as has been previously assumed; still, there are properties of individual structure that the aggregate networks do not reflect. A simulation of the new sampling process suggests that it can uncover the true structure of an individual’s semantic memory. (shrink)

While philosophers of language have traditionally relied upon their intuitions about cases when developing theories of reference, this methodology has recently been attacked on the grounds that intuitions about reference, far from being universal, show significant cultural variation, thus undermining their relevance for semantic theory. I’ll attempt to demonstrate that (1) such criticisms do not, in fact, undermine the traditional philosophical methodology, and (2) our underlying intuitions about the nature of reference may be more universal than the authors suppose.

Frederick Suppe has come to enjoy a position of undisputed leadership in the post-positivistic philosophy of science.

In his 1982 book Wittgenstein on Rules and Private Language, Saul Kripke develops a famous argument that purports to show that there are no facts about what we mean by the expressions of our language: ascriptions of meaning, such as “Jones means addition by ‘+’” or Smith means green by ‘green’”, are according to Kripke’s Wittgenstein neither true nor false. Kripke’s Wittgenstein thus argues for a form of non-factualism about ascriptions of meaning: ascriptions of meaning do not purport to state (...) facts. Define semantic realism to be the view that ascriptions of meaning are apt to be assessed in terms of truth and falsity, and are, at least in some instances, true. Semantic realism, thus defined, is a form of cognitivism about semantic judgement, according to which judgements ascribing meaning express beliefs, states apt for assessment in terms of truth and falsity. Kripke’s Wittgenstein thus argues against semantic realism, and in favour of a form of semantic non-cognitivism.1 En route to semantic non-cognitivism, Kripke’s Wittgenstein argues against dispositionalist theories of meaning, which hold something roughly along the lines of the following: Jones means magpie by “magpie” if and only if Jones is disposed to apply “magpie” to magpies in ideal conditions (where the ideal conditions can be specified in terms that don’t presuppose the notion of meaning, and are such that in those conditions Jones applies “magpie” to a thing if and only if that thing is a magpie). (shrink)

In this paper I show that we have strong empirical and theoretical reasons to treat the verbs we use in our semantic theorizing—particularly ‘refers to ’, ‘applies to ’, and ‘is true of ’—as intensional transitive verbs. Stating our semantic theories with intensional vocabulary allows us to partially reconcile two competing approaches to the nature and subject-matter of semantics: the Chomskian approach, on which semantics is non-relational, internalistic, and concerns the psychology of language users, and the Lewisian approach, (...) on which semantics is fully relational, specifies truth-conditions, and has metaphysical implications. ITVs have two readings: an intensional, de dicto reading, and a relational, de re reading. A semantic theory stated with the de dicto readings of our semantic verbs captures the core insights of the Chomskian approach to semantics, in part because it allows us to assign extremely fine-grained semantic values to expressions, even when those expressions are empty. On the other hand, the de re reading yields a theory that is fully relational, and issues in truth-conditions. The resulting theories are related—and compatible—in that they are expressed by two different readings of the very same semantic vocabulary, and plausibly, the distinction between these two readings is one of scope. (shrink)

The associative theory of creativity states that creativity is associated with differences in the structure of semantic memory, whereas the executive theory of creativity emphasises the role of top-down control for creative thought. For a powerful test of these accounts, individual semantic memory structure was modelled with a novel method based on semantic relatedness judgements and different criteria for network filtering were compared. The executive account was supported by a correlation between creative ability and broad retrieval ability. (...) The associative account was independently supported, when network filtering was based on a relatedness threshold, but not when it was based on a fixed edge number or on the analysis of weighted networks. In the former case, creative ability was associated with shorter average path lengths and higher clustering of the network, suggesting that the semantic networks of creative people show higher small-worldness. (shrink)

In this paper, we do two things. First, we provide some support for adopting a version of the meaningless strategy with respect to the liar paradox, and, second, we extend that strategy, by providing, albeit tentatively, a solution to that paradox—one that is semantic, rather than logical.

In this paper, we defend a traditional approach to semantics, that holds that the outputs of compositional semantics are propositional, i.e. truth conditions. Though traditional, this view has been challenged on a number of fronts over the years. Since classic work of Lewis, arguments have been offered which purport to show that semantic composition requires values that are relativized, e.g. to times, or other parameters that render them no longer propositional. Focusing in recent variants of these arguments involving quantification (...) and binding, we argue that a correct understanding of how composition works gives no reason to relativize semantic values, and that propositional semantic values are in fact the preferred option. We take our argument to be mainly empirical, but along the way, we defend some more general theses. Simple propositional semantic values are viable in composition, we maintain, because composition is itself a complex phenomenon, involving multiple modes of composition. Furthermore, some composition principles make adjustments to the meanings of constituents in the course of composition. These adjustments are by triggered syntactic environments. We argue such small contributions of meaning from syntactic structure are acceptable. (shrink)

Philosophers and cognitive scientists have worried that research on animal mind-reading faces a ‘logical problem’: the difficulty of experimentally determining whether animals represent mental states (e.g. seeing) or merely the observable evidence (e.g. line-of-gaze) for those mental states. The most impressive attempt to confront this problem has been mounted recently by Robert Lurz. However, Lurz' approach faces its own logical problem, revealing this challenge to be a special case of the more general problem of distal content. Moreover, participants in this (...) debate do not agree on criteria for representation. As such, future debate should either abandon the representational idiom or confront underlying semantic disagreements. (shrink)