PAGE 8

 The representationalist thesis is often formulated as the claim that phenomenal properties are representational or intentional. However, this formulation is ambiguous between a reductive and a non-deductive claim (though the term 'representationalism' is most often used for the reductive claim). On one hand, it could mean that the phenomenal content of an experience is a kind of intentional content (the properties it represents). On the other, it could mean that the (irreducible) phenomenal properties of an experience determine an intentional content. Representationalists such as Dretske, Lycan and Tye would assent to the former claim, whereas phenomenalists such as Block, Chalmers, Loar and Peacocke would assent to the latter. (Among phenomenalists, there is further disagreement about whether Qualia are intrinsically representational (Loar) or not (Block, Peacocke).
 Most (reductive) representationalists are motivated by the conviction that one or another naturalistic explanation of intentionality is, in broad outline, correct, and by the desire to complete the naturalization of the mental by applying such theories to the problem of phenomenality. (Needless to say, most phenomenalists (Chalmers is the major exception) are just as eager to naturalize the phenomenal - though not in the same way.)
 The main argument for representationalism appeals to the transparency of experience. The properties that characterize what it's like to have a perceptual experience are presented in experience as properties of objects perceived: in attending to an experience, one seems to 'see through it' to the objects and properties it is experiences of. They are not presented as properties of the experience itself. If nonetheless they were properties of the experience, perception would be massively deceptive. But perception is not massively deceptive. According to the representationalist, the phenomenal character of an experience is due to its representing objective, non-experiential properties. (In veridical perception, these properties are locally instantiated; in illusion and hallucination, they are not.) On this view, introspection is indirect perception: one comes to know what phenomenal features one's experience has by coming to know what objective features it represents.
 In order to account for the intuitive differences between conceptual and sensory representations, representationalists appeal to their structural or functional differences. Dretske (1995), for example, distinguishes experiences and thoughts on the basis of the origin and nature of their functions: an experience of a property 'P' is a state of a system whose evolved function is to indicate the presence of 'P' in the environment; a thought representing the property 'P', on the other hand, is a state of a system whose assigned (learned) function is to calibrate the output of the experiential system. Rey (1991) takes both thoughts and experiences to be relations to sentences in the language of thought, and distinguishes them on the basis of (the functional roles of) such sentences' constituent predicates. Lycan (1987, 1996) distinguishes them in terms of their functional-computational profiles. Tye (2000) distinguishes them in terms of their functional roles and the intrinsic structure of their vehicles: thoughts are representations in a language-like medium, whereas experiences are image-like representations consisting of 'symbol-filled arrays.' (The account of mental images in Tye 1991.)
 Phenomenalists tend to make use of the same sorts of features (function, intrinsic structure) in explaining some of the intuitive differences between thoughts and experiences; but they do not suppose that such features exhaust the differences between phenomenal and non-phenomenal representations. For the phenomenalist, it is the phenomenal properties of experiences - Qualia themselves - that constitute the fundamental difference between experience and thought. Peacocke (1992), for example, develops the notion of a perceptual 'scenario' (an assignment of phenomenal properties to coordinates of a three-dimensional egocentric space), whose content is 'correct' (a semantic property) if in the corresponding 'scene' (the portion of the external world represented by the scenario) properties are distributed as their phenomenal analogues are in the scenario.
 Another sort of representation championed by phenomenalists (e.g., Block, Chalmers (2003) and Loar (1996)) is the 'phenomenal concept' -, a conceptual/phenomenal hybrid consisting of a phenomenological 'sample' (an image or an occurrent sensation) integrated with (or functioning as) a conceptual component. Phenomenal concepts are postulated to account for the apparent fact (among others) that, as McGinn (1991) puts it, 'you cannot form [introspective] concepts of conscious properties unless you yourself instantiate those properties.' One cannot have a phenomenal concept of a phenomenal property 'P', and, hence, phenomenal beliefs about P, without having experience of 'P', because 'P' itself is (in some way) constitutive of the concept of 'P'. (Jackson 1982, 1986 and Nagel 1974.)
 Though imagery has played an important role in the history of philosophy of mind, the important contemporary literature on it is primarily psychological. In a series of psychological experiments done in the 1970s (summarized in Kosslyn 1980 and Shepard and Cooper 1982), subjects' response time in tasks involving mental manipulation and examination of presented figures was found to vary in proportion to the spatial properties (size, orientation, etc.) of the figures presented. The question of how these experimental results are to be explained has kindled a lively debate on the nature of imagery and imagination.
 Kosslyn (1980) claims that the results suggest that the tasks were accomplished via the examination and manipulation of mental representations that they have spatial properties, i.e., pictorial representations, or images. Others, principally Pylyshyn (1979, 1981, 2003), argue that the empirical facts can be explained in terms exclusively of discursive, or propositional representations and cognitive processes defined over them. (Pylyshyn takes such representations to be sentences in a language of thought.)
 The idea that pictorial representations are literally pictures in the head is not taken seriously by proponents of the pictorial view of imagery.  The claim is, rather, that mental images represent in a way that is relevantly like the way pictures represent. (Attention has been focussed on visual imagery - hence the designation 'pictorial'; Though of course, there may imagery in other modalities - auditory, olfactory, etc. - as well.)
 The distinction between pictorial and discursive representation can be characterized in terms of the distinction between analog and digital representation (Goodman 1976). This distinction has itself been variously understood (Fodor & Pylyshyn 1981, Goodman 1976, Haugeland 1981, Lewis 1971, McGinn 1989), though a widely accepted construal is that analog representation is continuous (i.e., in virtue of continuously variable properties of the representation), while digital representation is discrete (i.e., in virtue of properties a representation either has or doesn't have) (Dretske 1981). (An analog/digital distinction may also be made with respect to cognitive processes. (Block 1983.)) On this understanding of the analog/digital distinction, imaginistic representations, which represent in virtue of properties that may vary continuously (such for being more or less bright, loud, vivid, etc.), would be analog, while conceptual representations, whose properties do not vary continuously (a thought cannot be more or less about Elvis: either it is or it is not) would be digital.
 It might be supposed that the pictorial/discursive distinction is best made in terms of the phenomenal/nonphenomenal distinction, but it is not obvious that this is the case. For one thing, there may be nonphenomenal properties of representations that vary continuously. Moreover, there are ways of understanding pictorial representation that presuppose neither phenomenality nor analogicity. According to Kosslyn (1980, 1982, 1983), a mental representation is 'quasi-pictorial' when every part of the representation corresponds to a part of the object represented, and relative distances between parts of the object represented are preserved among the parts of the representation. But distances between parts of a representation can be defined functionally rather than spatially - for example, in terms of the number of discrete computational steps required to combine stored information about them. (Rey 1981.)
 Tye (1991) proposes a view of images on which they are hybrid representations, consisting both of pictorial and discursive elements. On Tye's account, images are '(labelled) interpreted symbol-filled arrays.' The symbols represent discursively, while their arrangement in arrays has representational significance (the location of each 'cell' in the array represents a specific viewer-centred 2-D location on the surface of the imagined object)
 The contents of mental representations are typically taken to be abstract objects (properties, relations, propositions, sets, etc.). A pressing question, especially for the naturalist, is how mental representations come to have their contents. Here the issue is not how to naturalize content (abstract objects can't be naturalized), but, rather, how to provide a naturalistic account of the content-determining relations between mental representations and the abstract objects they express. There are two basic types of contemporary naturalistic theories of content-determination, causal-informational and functional.
 Causal-informational theories hold that the content of a mental representation is grounded in the information it carries about what does (Devitt 1996) or would (Fodor 1987, 1990) cause it to occur. There is, however, widespread agreement that causal-informational relations are not sufficient to determine the content of mental representations. Such relations are common, but representation is not. Tree trunks, smoke, thermostats and ringing telephones carry information about what they are causally related to, but they do not represent (in the relevant sense) what they carry information about. Further, a representation can be caused by something it does not represent, and can represent something that has not caused it.
 The main attempts to specify what makes a causal-informational state a mental representation are Asymmetric Dependency Theories, the Asymmetric Dependency Theory distinguishes merely informational relations from representational relations on the basis of their higher-order relations to each other: informational relations depend upon representational relations, but not vice-versa. For example, if tokens of a mental state type are reliably caused by horses, cows-on-dark-nights, zebras-in-the-mist and Great Danes, then they carry information about horses, etc. If, however, such tokens are caused by cows-on-dark-nights, etc. because they were caused by horses, but not vice versa, then they represent horses.
 According to Teleological Theories, representational relations are those a representation-producing mechanism has the selected (by evolution or learning) function of establishing. For example, zebra-caused horse-representations do not mean zebra, because the mechanism by which such tokens are produced has the selected function of indicating horses, not zebras. The horse-representation-producing mechanism that responds to zebras is malfunctioning.
 Functional theories, hold that the content of a mental representation are well  grounded in causal computational inferential relations to other mental portrayals other than mental representations.  They differ on whether relata should include all other mental representations or only some of them, and on whether to include external states of affairs. The view that the content of a mental representation is determined by its inferential/computational relations with all other representations is holism; the view it is determined by relations to only some other mental states is localisms (or molecularism). (The view that the content of a mental state depends on none of its relations to other mental states is atomism.) Functional theories that recognize no content-determining external relata have been called solipsistic (Harman 1987). Some theorists posit distinct roles for internal and external connections, the former determining semantic properties analogous to sense, the latter determining semantic properties analogous to reference (McGinn 1982, Sterelny 1989)
 (Reductive) representationalists (Dretske, Lycan, Tye) usually take one or another of these theories to provide an explanation of the (Non-conceptual) content of experiential states. They thus tend to be Externalists, about phenomenological as well as conceptual content. Phenomenalists and non-deductive representationalists (Block, Chalmers, Loar, Peacocke, Siewert), on the other hand, take it that the representational content of such states is (at least in part) determined by their intrinsic phenomenal properties. Further, those who advocate a phenomenology-based approach to conceptual content (Horgan and Tiensen, Loar, Pitt, Searle, Siewert) also seem to be committed to Internalist individuation of the content (if not the reference) of such states.
 Generally, those who, like informational theorists, think relations to one's (natural or social) environment are (at least partially) determinative of the content of mental representations are Externalists (e.g., Burge 1979, 1986, McGinn 1977, Putnam 1975), whereas those who, like some proponents of functional theories, think representational content is determined by an individual's intrinsic properties alone, are internalists (or individualists).
 This issue is widely taken to be of central importance, since psychological explanation, whether commonsense or scientific, is supposed to be both causal and content-based. (Beliefs and desires cause the behaviours they do because they have the contents they do. For example, the desire that one have a beer and the beliefs that there is beer in the refrigerator and that the refrigerator is in the kitchen may explain one's getting up and going to the kitchen.) If, however, a mental representation's having a particular content is due to factors extrinsic to it, it is unclear how its having that content could determine its causal powers, which, arguably, must be intrinsic. Some who accept the standard arguments for externalism have argued that internal factors determine a component of the content of a mental representation. They say that mental representations have both 'narrow' content (determined by intrinsic factors) and 'wide' or 'broad' content (determined by narrow content plus extrinsic factors). (This distinction may be applied to the sub-personal representations of cognitive science as well as to those of commonsense psychology.
 Narrow content has been variously construed. Putnam (1975), Fodor (1982)), and Block (1986) for example, seems to understand it as something like dedictorial content (i.e., Frégean sense, or perhaps character, à la Kaplan 1989). On this construal, narrow content is context-independent and directly expressible. Fodor (1987) and Block (1986), however, has also characterized narrow content as radically inexpressible. On this construal, narrow content is a kind of proto-content, or content-determinant, and can be specified only indirectly, via specifications of context/wide-content pairings. Both, construe of as a narrow content and are characterized as functions from context to (wide) content. The narrow content of a representation is determined by properties intrinsic to it or its possessor such as its syntactic structure or its intra-mental computational or inferential role or its phenomenology.
 Burge (1986) has argued that causation-based worries about externalist individuation of psychological content, and the introduction of the narrow notion, are misguided. Fodor (1994, 1998) has more recently urged that there may be no need to narrow its contentual representations, accountable for reasons of an ordering supply of naturalistic (causal) explanations of human cognition and action, since the sorts of cases they were introduced to handle, viz., Twin-Earth cases and Frége cases, are nomologically either impossible or dismissible as exceptions to non-strict psychological laws.
 The leading contemporary version of the Representational Theory of Mind, the Computational Theory of Mind, claims that the brain is a kind of computer and that mental processes are computations. According to the computational theory of mind, cognitive states are constituted by computational relations to mental representations of various kinds, and cognitive processes are sequences of such states. The computational theory of mind and the representational theory of mind, may by attempting to explain all psychological states and processes in terms of mental representation. In the course of constructing detailed empirical theories of human and animal cognition and developing models of cognitive processes' implementable in artificial information processing systems, cognitive scientists have proposed a variety of types of mental representations. While some of these may be suited to be mental relata of commonsense psychological states, some - so-called 'subpersonal' or 'sub-doxastic' representations - are not. Though many philosophers believe that computational theory of mind can provide the best scientific explanations of cognition and behaviour, there is disagreement over whether such explanations will vindicate the commonsense psychological explanations of prescientific representational theory of mind.
 According to Stich's (1983) Syntactic Theory of Mind, for example, computational theories of psychological states should concern themselves only with the formal properties of the objects those states are relations to. Commitment to the explanatory relevance of content, however, is for most cognitive scientists fundamental. That mental processes are computations, which computations are rule-governed sequences of semantically evaluable objects, and that the rules apply to the symbols in virtue of their content, are central tenets of mainstream cognitive science.
 Explanations in cognitive science appeal to a many different kinds of mental representation, including, for example, the 'mental models' of Johnson-Laird 1983, the 'retinal arrays,' 'primal sketches' and '2½ -D sketches' of Marr 1982, the 'frames' of Minsky 1974, the 'sub-symbolic' structures of Smolensky 1989, the 'quasi-pictures' of Kosslyn 1980, and the 'interpreted symbol-filled arrays' of Tye 1991 - in addition to representations that may be appropriate to the explanation of commonsense
Psychological states. Computational explanations have been offered of, among other mental phenomena, belief.
 The classicists hold that mental representations are symbolic structures, which typically have semantically evaluable constituents, and that mental processes are rule-governed manipulations of them that are sensitive to their constituent structure. The connectionists, hold that mental representations are realized by patterns of activation in a network of simple processors ('nodes') and that mental processes consist of the spreading activation of such patterns. The nodes themselves are, typically, not taken to be semantically evaluable; nor do the patterns have semantically evaluable constituents. (Though there are versions of Connectionism -, 'localist' versions - on which individual nodes are taken to have semantic properties (e.g., Ballard 1986, Ballard & Hayes 1984).) It is arguable, however, that localist theories are neither definitive nor representative of the Conceptionist program.
 Classicists are motivated (in part) by properties thought seems to share with language. Jerry Alan Fodor's (1935-), Language of Thought Hypothesis, (Fodor 1975, 1987), according to which the system of mental symbols constituting the neural basis of thought is structured like a language, provides a well-worked-out version of the classical approach as applied to commonsense psychology. According to the language of a thought hypothesis, the potential infinity of complex representational mental states is generated from a finite stock of primitive representational states, in accordance with recursive formation rules. This combinatorial structure accounts for the properties of productivity and systematicity of the system of mental representations. As in the case of symbolic languages, including natural languages (though Fodor does not suppose either that the language of thought hypotheses explains only linguistic capacities or that only verbal creatures have this sort of cognitive architecture), these properties of thought are explained by appeal to the content of the representational units and their combinability into contentful complexes. That is, the semantics of both language and thought is compositional: the content of a complex representation is determined by the contents of its constituents and their structural configuration.
 Connectionists are motivated mainly by a consideration of the architecture of the brain, which apparently consists of layered networks of interconnected neurons. They argue that this sort of architecture is unsuited to carrying out classical serial computations. For one thing, processing in the brain is typically massively parallel. In addition, the elements whose manipulation drive's computation in Conceptionist networks (principally, the connections between nodes) are neither semantically compositional nor semantically evaluable, as they are on the classical approach. This contrast with classical computationalism is often characterized by saying that representation is, with respect to computation, distributed as opposed to local: representation is local if it is computationally basic; and distributed if it is not. (Another way of putting this is to say that for classicists mental representations are computationally atomic, whereas for connectionists they are not.)
 Moreover, connectionists argue that information processing as it occurs in Conceptionist networks more closely resembles some features of actual human cognitive functioning. For example, whereas on the classical view learning involves something like hypothesis formation and testing (Fodor 1981), on the Conceptionist model it is a matter of evolving distribution of 'weight' (strength) on the connections between nodes, and typically does not involve the formulation of hypotheses regarding the identity conditions for the objects of knowledge. The Conceptionist network is 'trained up' by repeated exposure to the objects it is to learn to distinguish; and, though networks typically require many more exposures to the objects than do humans, this seems to model at least one feature of this type of human learning quite well.
 Further, degradation in the performance of such networks in response to damage is gradual, not sudden as in the case of a classical information processor, and hence more accurately models the loss of human cognitive function as it typically occurs in response to brain damage. It is also sometimes claimed that Conceptionist systems show the kind of flexibility in response to novel situations typical of human cognition - situations in which classical systems are relatively 'brittle' or 'fragile.'
 Some philosophers have maintained that Connectionism entails that there are no propositional attitudes. Ramsey, Stich and Garon (1990) have argued that if Conceptionist models of cognition are basically correct, then there are no discrete representational states as conceived in ordinary commonsense psychology and classical cognitive science. Others, however (e.g., Smolensky 1989), hold that certain types of higher-level patterns of activity in a neural network may be roughly identified with the representational states of commonsense psychology. Still others argue that language-of-thought style representation is both necessary in general and realizable within Conceptionist architectures, collect the central contemporary papers in the classicist/Conceptionist debate, and provides useful introductory material as well.
 Whereas Stich (1983) accepts that mental processes are computational, but denies that computations are sequences of mental representations, others accept the notion of mental representation, but deny that computational theory of mind provides the correct account of mental states and processes.
 Van Gelder (1995) denies that psychological processes are computational. He argues that cognitive systems are dynamic, and that cognitive states are not relations to mental symbols, but quantifiable states of a complex system consisting of (in the case of human beings) a nervous system, a body and the environment in which they are embedded. Cognitive processes are not rule-governed sequences of discrete symbolic states, but continuous, evolving total states of dynamic systems determined by continuous, simultaneous and mutually determining states of the systems components. Representation in a dynamic system is essentially information-theoretic, though the bearers of information are not symbols, but state variables or parameters.
 Horst (1996), on the other hand, argues that though computational models may be useful in scientific psychology, they are of no help in achieving a philosophical understanding of the intentionality of commonsense mental states. Computational theory of mind attempts to reduce the intentionality of such states to the intentionality of the mental symbols they are relations to. But, Horst claims, the relevant notion of symbolic content is essentially bound up with the notions of convention and intention. So the computational theory of mind involves itself in a vicious circularity: the very properties that are supposed to be reduced are (tacitly) appealed to in the reduction.
 To say that a mental object has semantic properties is, paradigmatically, to say that it may be about, or be true or false of, an object or objects, or that it may be true or false simpiciter. Suppose I think that you took to sniffing snuff. I am thinking about you, and if what I think of you (that they take snuff) is true of you, then my thought is true. According to representational theory of mind such states are to be explained as relations between agents and mental representations. To think that you take snuff is to token in some way a mental representation whose content is that ocelots take snuff. On this view, the semantic properties of mental states are the semantic properties of the representations they are relations to.
 Linguistic acts seem to share such properties with mental states. Suppose I say that you take snuff. I am talking about you, and if what I say of you (that they take snuff) is true of them, then my utterance is true. Now, to say that you take snuff is (in part) to utter a sentence that means that you take snuff. Many philosophers have thought that the semantic properties of linguistic expressions are inherited from the intentional mental states they are conventionally used to express. On this view, the semantic properties of linguistic expressions are the semantic properties of the representations that are the mental relata of the states they are conventionally used to express.
 It is also widely held that in addition to having such properties as reference, truth-conditions and truth - so-called extensional properties - expressions of natural languages also have intensional properties, in virtue of expressing properties or propositions - i.e., in virtue of having meanings or senses, where two expressions may have the same reference, truth-conditions or truth value, yet express different properties or propositions (Frége 1892/1997). If the semantic properties of natural-language expressions are inherited from the thoughts and concepts they express (or vice versa, or both), then an analogous distinction may be appropriate for mental representations.
 Theories of representational content may be classified according to whether they are atomistic or holistic and according to whether they are externalistic or internalistic, whereby, emphasizing the priority of a whole over its parts. Furthermore, in the philosophy of language, this becomes the claim that the meaning of an individual word or sentence can only be understood in terms of its relation to an indefinitely larger body of language, such as à whole theory, or even a whole language or form of life. In the philosophy of mind a mental state similarly may be identified only in terms of its relations with others. Moderate holism may allow the other things besides these relationships also count; extreme holism would hold that a network of relationships is all that we have. A holistic view of science holds that experience only confirms or disconfirms large bodies of doctrine, impinging at the edges, and leaving some leeway over the adjustment that it requires.
 Once, again, in the philosophy of mind and language, the view that what is thought, or said, or experienced, is essentially dependent on aspects of the world external to the mind of the subject. The view goes beyond holding that such mental states are typically caused by external factors, to insist that they could not have existed as they now do without the subject being embedded in an external world of a certain kind. It is these external relations that make up the essence or identify of the mental state. Externalism is thus opposed to the Cartesian separation of the mental from the physical, since that holds that the mental could in principle exist as it does even if there were no external world at all. Various external factors have been advanced as ones on which mental content depends, including the usage of experts, the linguistic, norms of the community. And the general causal relationships of the subject. In the theory of knowledge, externalism is the view that a person might know something by being suitably situated with respect to it, without that relationship being in any sense within his purview. The person might, for example, be very reliable in some respect without believing that he is. The view allows that you can know without being justified in believing that you know.
 However, atomistic theories take a representation's content to be something that can be specified independent entity of that representation' s relations to other representations. What the American philosopher of mind, Jerry Alan Fodor (1935-) calls the crude causal theory, for example, takes a representation to be a |cow| - a menial representation with the same content as the word 'cow' - if its tokens are caused by instantiations of the property of being-a-cow, and this is a condition that places no explicit constraints on how |cow|'s must or might relate to other representations. Holistic theories contrasted with atomistic theories in taking the relations à representation bears to others to be essential to its content. According to functional role theories, a representation is a |cow| if it behaves like a |cow| should behave in inference.
 Internalist theories take the content of a representation to be a matter determined by factors internal to the system that uses it. Thus, what Block (1986) calls 'short-armed' functional role theories are Internalist. Externalist theories take the content of a representation to be determined, in part at least, by factors external to the system that uses it. Covariance theories, as well as teleological theories that invoke an historical theory of functions, take content to be determined by 'external' factors. Crossing the atomist-holistic distinction with the Internalist-externalist distinction.
 Externalist theories (sometimes called non-individualistic theories) have the consequence that molecule for molecule identical cognitive systems might yet harbour representations with different contents. This has given rise to a controversy concerning 'narrow' content. If we assume some form of externalist theory is correct, then content is, in the first instance 'wide' content, i.e., determined in part by factors external to the representing system. On the other hand, it seems clear that, on plausible assumptions about how to individuate psychological capacities, internally equivalent systems must have the same psychological capacities. Hence, it would appear that wide content cannot be relevant to characterizing psychological equivalence. Since cognitive science generally assumes that content is relevant to characterizing psychological equivalence, philosophers attracted to externalist theories of content have sometimes attempted to introduce 'narrow' content, i.e., an aspect or kind of content that is equivalent internally equivalent systems. The simplest such theory is Fodor's idea (1987) that narrow content is a function from contents (i.e., from whatever the external factors are) to wide contents.
 All the same, what a person expresses by a sentence is often a function of the environment in which he or she is placed. For example, the disease I refer to by the term like 'arthritis', or the kind of tree I refer to as a 'Maple' will be defined by criteria of which I know next to nothing. This raises the possibility of imagining two persons in rather different environments, but in which everything appears the same to each of them. The wide content of their thoughts and sayings will be different if the situation surrounding them is appropriately different: 'situation' may include the actual objects they perceive or the chemical or physical kinds of object in the world they inhabit, or the history of their words, or the decisions of authorities on what counts as an example, of one of the terms they use. The narrow content is that part of their thought which remains identical, through their identity of the way things appear, regardless of these differences of surroundings. Partisans of wide content may doubt whether any content in this sense narrow, partisans of narrow content believer that it is the fundamental notion, with wide content being explicable in terms of narrow content plus context.
 Even so, the distinction between facts and values has outgrown its name: it applies not only to matters of fact vs, matters of value, but also to statements that something is, vs. statements that something ought to be. Roughly, factual statements - 'is statements' in the relevant sense - represent some state of affairs as obtaining, whereas normative statements - evaluative, and deontic ones - attribute goodness to something, or ascribe, to an agent, an obligation to act. Neither distinction is merely linguistic. Specifying a book's monetary value is making a factual statement, though it attributes a kind of value. 'That is a good book' expresses a value judgement though the term 'value' is absent (nor would 'valuable' be synonymous with 'good'). Similarly, 'we are morally obligated to fight' superficially expresses a statement, and 'By all indications it ought to rain' makes a kind of ought-claim; but the former is an ought-statement, the latter an (epistemic) is-statement.
 Theoretical difficulties also beset the distinction. Some have absorbed values into facts holding that all value is instrumental, roughly, to have value is to contribute - in a factual analysable way - to something further which is (say) deemed desirable. Others have suffused facts with values, arguing that facts (and observations) are 'theory-impregnated' and contending that values are inescapable to theoretical choice. But while some philosophers doubt that fact/value distinctions can be sustained, there persists a sense of a deep difference between evaluating, and attributing an obligation and, on the other hand, saying how the world is.
 Fact/value distinctions, may be defended by appeal to the notion of intrinsic value, as a thing has in itself and thus independently of its consequences. Roughly, a value statement (proper) is an ascription of intrinsic value, one to the effect that a thing is to some degree good in itself. This leaves open whether ought-statements are implicitly value statements, but even if they imply that something has intrinsic value - e.g., moral value - they can be independently characterized, say by appeal to rules that provide (justifying) reasons for action. One might also ground the fact value distinction in the attributional (or even motivational) component apparently implied by the making of valuational or deontic judgements: Thus, 'it is a good book, but that is no reason for a positive attribute towards it' and 'you ought to do it, but there is no reason to' seem inadmissible, whereas, substituting, 'an expensive book' and 'you will do it' yields permissible judgements. One might also argue that factual judgements are the kind which are in principle appraisable scientifically, and thereby anchor the distinction on the factual side. This ligne is plausible, but there is controversy over whether scientific procedures are 'value-free' in the required way.
 Philosophers differ regarding the sense, if any, in which epistemology is normative (roughly, valuational). But what precisely is at stake in this controversy is no clearly than the problematic fact/value distinction itself. Must epistemologists as such make judgements of value or epistemic responsibility? If epistemology is naturalizable, then even epistemic principles simply articulate under what conditions - say, appropriate perceptual stimulations - a belief is justified, or constitutes knowledge. Its standards of justification, then would be like standards of, e.g., resilience for bridges. It is not obvious, however, that there appropriate standards can be established without independent judgements that, say, a certain kind of evidence is good enough for justified belief (or knowledge). The most plausible view may be that justification is like intrinsic goodness, though it supervenes on natural properties, it cannot be analysed wholly in factual statements.
 Thus far, belief has been depicted as being all-or-nothing, however, as a resulting causality for which we have grounds for thinking it true, and, all the same, its acceptance is governed by epistemic norms, and, least of mention, it is partially subject to voluntary control and has functional affinities to belief. Still, the notion of acceptance, like that of degrees of belief, merely extends the standard picture, and does not replace it.
 Traditionally, belief has been of epistemological interest in its propositional guise: 'S' believes that 'p', where 'p' is a reposition towards which an agent, 'S' exhibits an attitude of acceptance. Not all belief is of this sort. If I trust you to say, I believer you. And someone may believe in Mr. Radek, or in a free-market economy, or in God. It is sometimes supposed that all belief is 'reducible' to propositional belief, belief-that. Thus, my believing you might be thought a matter of my believing, is,  perhaps, that what you say is true, and your belief in free markets or God, is a matter of your believing that free-market economies are desirable or that God exists.
 Some philosophers have followed St. Thomas Aquinas (1225-74), in supposing that to believer in God is simply to believer that certain truths hold while others argue that belief-in is a distinctive attitude, on that includes essentially an element of trust. More commonly, belief-in has been taken to involve a combination of propositional belief together with some further attitude.
 The moral philosopher Richard Price (1723-91) defends the claim that there are different sorts of belief-in, some, but not all reducible to beliefs-that. If you believer in God, you believer that God exists, that God is good, you believer that God is good, etc. But according to Price, your belief involves, in addition, a certain complex pro-attitude toward its object. Even so, belief-in outruns the evidence for the corresponding belief-that. Does this diminish its rationality? If belief-in presupposes believes-that, it might be thought that the evidential standards for the former must be, at least, as high as standards for the latter. And any additional pro-attitude might be thought to require a further layer of justification not required for cases of belief-that.
 Belief-in may be, in general, less susceptible to alternations in the face of unfavourable evidence than belief-that. A believer who encounters evidence against God's existence may remain unshaken in his belief, in part because the evidence does not bear on his pro-attitude. So long as this ids united with his belief that God exists, and reasonably so - in a way that an ordinary propositional belief that would not.
 The correlative way of elaborating on the general objection to justificatory externalism challenges the sufficiency of the various externalist conditions by citing cases where those conditions are satisfied, but where the believers in question seem intuitively not to be justified. In this context, the most widely discussed examples have to do with possible occult cognitive capacities, like clairvoyance. Considering the point in application once, again, to reliabilism, the claim is that to think that he has such a cognitive power, and, perhaps, even good reasons to the contrary, is not rational or responsible and therefore not epistemically justified in accepting the belief that result from his clairvoyance, despite the fact that the reliabilist condition is satisfied.
 One sort of response to this latter sorts of an objection is to 'bite the bullet' and insist that such believers are in fact justified, dismissing the seeming intuitions to the contrary as latent Internalist prejudice. A more widely adopted response attempts to impose additional conditions, usually of a roughly Internalist sort, which will rule out the offending example, while stopping far of a full internalism. But, while there is little doubt that such modified versions of externalism can handle particular cases, as well enough to avoid clear intuitive implausibility, the usually problematic cases that they cannot handle, and also whether there is and clear motivation for the additional requirements other than the general Internalist view of justification that externalist is committed to reject.
 A view in this same general vein, one that might be described as a hybrid of internalism and externalism holds that epistemic justification requires that there is a justicatory factor that is cognitively accessible to the believer in question (though it need not be actually grasped), thus ruling out, e.g., a pure reliabilism. At the same time, however, though it must be objectively true that beliefs for which such a factor is available are likely to be true, in addition, the fact need not be in any way grasped or cognitively accessible to the believer. In effect, of the premises needed to argue that a particular belief is likely to be true, one must be accessible in a way that would satisfy at least weak internalism, the Internalist will respond that this hybrid view is of no help at all in meeting the objection and has no belief nor is it held in the rational, responsible way that justification intuitively seems to require, for the believer in question, lacking one crucial premise, still has no reason at all for thinking that his belief is likely to be true.
 An alternative to giving an externalist account of epistemic justification, one which may be more defensible while still accommodating many of the same motivating concerns, is to give an externalist account of knowledge directly, without relying on an intermediate account of justification. Such a view will obviously have to reject the justified true belief account of knowledge, holding instead that knowledge is true belief which satisfies the chosen externalist condition, e.g., a result of a reliable process (and perhaps, further conditions as well). This makes it possible for such a view to retain Internalist account of epistemic justification, though the centrality of that concept to epistemology would obviously be seriously diminished.
 Such an externalist account of knowledge can accommodate the commonsense conviction that animals, young children, and unsophisticated adults' posse's knowledge, though not the weaker conviction (if such a conviction does exist) that such individuals are epistemically justified in their beliefs. It is, at least, less vulnerable to Internalist counter-examples of the sort discussed, since the intuitions involved there pertain more clearly to justification than to knowledge. What is uncertain is what ultimate philosophical significance the resulting conception of knowledge, for which is accepted or advanced as true or real on the basis of less than conclusive evidence, as can only be assumed to have. In particular, does it have any serious bearing on traditional epistemological problems and on the deepest and most troubling versions of scepticism, which seems in fact to be primarily concerned with justification, and knowledge?`
 A rather different use of the terms 'internalism' and 'externalism' have to do with the issue of how the content of beliefs and thoughts is determined: According to an Internalist view of content, the content of such intention states depends only on the non-relational, internal properties of the individual's mind or grain, and not at all on his physical and social environment: While according to an externalist view, content is significantly affected by such external factors and suggests a view that appears of both internal and external elements are standardly classified as an external view.
 As with justification and knowledge, the traditional view of content has been strongly Internalist in character. The main argument for externalism derives from the philosophy y of language, more specifically from the various phenomena pertaining to natural kind terms, indexicals, etc. that motivate the views that have come to be known as 'direct reference' theories. Such phenomena seem at least to show that the belief or thought content that can be properly attributed to a person is dependant on facts about his environment, e.g., whether he is on Earth or Twin Earth, what is fact pointing at, the classificatory criterion employed by expects in his social group, etc. - not just on what is going on internally in his mind or brain.
 An objection to externalist account of content is that they seem unable to do justice to our ability to know the content of our beliefs or thought 'from the inside', simply by reflection. If content is depending on external factors pertaining to the environment, then knowledge of content should depend on knowledge of these factors - which will not in general be available to the person whose belief or thought is in question.
 The adoption of an externalist account of mental content would seem to support an externalist account of justification, apart from all contentful representation is a belief inaccessible to the believer, then both the justifying statuses of other beliefs in relation to that of the same representation are the status of that content, being totally rationalized by further beliefs for which it will be similarly inaccessible. Thus, contravening the Internalist requirement for justification, as an Internalist must insist that there are no justification relations of these sorts, that our internally associable content can also not be warranted or as stated or indicated without the deviated departure from a course or procedure or from a norm or standard in showing no deviation from traditionally held methods of justification exacting by anything else: But such a response appears lame unless it is coupled with an attempt to show that the externalised account of content is mistaken.
 Except for alleged cases of thing s that are evident for one just by being true, it has often been thought, anything is known must satisfy certain criteria as well as being true. Except for alleged cases of self-evident truths, it is often thought that anything that is known must satisfy certain criteria or standards. These criteria are general principles that will make a proposition evident or just make accepting it warranted to some degree. Common suggestions for this role include position ‘p’, e.g., that 2 + 2 = 4, ‘p’ is evident or, if ‘p’ coheres wit h the bulk of one’s beliefs, ‘p’ is warranted. These might be criteria whereby putative self-evident truths, e.g., that one clearly and distinctly conceive s ‘p’, ‘transmit’ the status as evident they already have without criteria to other proposition s like ‘p’, or they might be criteria whereby purely non-epistemic considerations, e.g., facts about logical connections or about conception that need not be already evident or warranted, originally ‘create’ p’s epistemic status. If that in turn can be ‘transmitted’ to other propositions, e.g., by deduction or induction, there will be criteria specifying when it is.
 Nonetheless, of or relating to tradition a being previously characterized or specified to convey an idea indirectly, as an idea or theory for consideration and being so extreme a design or quality and lean towards an ecocatorial suggestion that implicate an involving responsibility that include: (1) if a proposition ‘p’, e.g., that 2 + 2 = 4, is clearly and distinctly conceived, then ‘p’ is evident, or simply, (2) if we can’t conceive ‘p’ to be false, then ‘p’ is evident: Or, (3) whenever are immediately conscious o f in thought or experience, e.g,, that we seem to see red, is evident. These might be criteria whereby putative self-evident truth s, e.g., that one clearly and distinctly conceives, e.g., that one clearly and distinctly conceives ‘p’, ‘transmit’ the status as evident they already have for one without criteria to other propositions like ‘p’. Alternatively, they might be criteria whereby epistemic status, e.g., p’s being evident, is originally created by purely non-epistemic considerations, e.g., facts about how ‘p’ is conceived which are neither self-evident is already criterial evident.
 The result effect, holds that traditional criteria do not seem to make evident propositions about anything beyond our own thoughts, experiences and necessary truths, to which deductive or inductive criteria ma y be applied. Moreover, arguably, inductive criteria, including criteria warranting the best explanation of data, never make things evident or warrant their acceptance enough to count as knowledge.
 Contemporary epistemologists suggest that traditional criteria may need alteration in three ways. Additional evidence may subject even our most basic judgements to rational correction, though they count as evident on the basis of our criteria. Warrant may be transmitted other than through deductive and inductive relations between propositions. Transmission criteria might not simply ‘pass’ evidence on linearly from a foundation of highly evident ‘premisses’ to ‘conclusions’ that are never more evident.
 A group of statements, some of which purportedly provide support for another. The statements which purportedly provide the support are the premisses while the statement purportedly support is the conclusion. Arguments are typically divided into two categories depending on the degree of support they purportedly provide. Deductive arguments purportedly provide conclusive support for their conclusions while inductively supports the purported provision that inductive arguments purportedly provided only arguments purportedly in the providing probably of support. Some, but not all, arguments succeed in providing support for their conclusions. Successful deductive arguments are valid while successful inductive arguments are valid while successful inductive arguments are strong. An argument is valid just in case if all its premisses are true its conclusion is only probably true. Deductive logic provides methods for ascertaining whether or not an argument is valid whereas, inductive logic provides methods for ascertaining the degree of support the premisses of an argument confer on its conclusion.
 Finally, proof, least of mention, is a collection of considerations and reasons that instill and sustain conviction that some proposed theorem - the theorem proved - is not only true, but could not possibly be false. A perceptual observation may instill the conviction that water is cold. But a proof that 2 + 5 = 5 must not only instill the conviction that is true that 2 + 3 = 5, but also that 2 + 3 could not be anything but 5.
 No one has succeeded in replacing this largely psychological characterization of proofs by a more objective characterization. The representations of reconstructions of proofs as mechanical and semiotical derivation in formal-logical systems all but completely fail to capture ‘proofs’ as mathematicians are quite content to give them. For example, formal-logical derivations depend solely on the logical form of the considered proposition, whereas usually proofs depend in large measure on content of propositions other than their logical form.
An alternative to giving an externalist account of epistemic justification, one which may be more defensible while still accommodating many of the same motivating concerns, is to give an externalist account of knowledge directly, without relying on an intermediate account of justification. Such a view will obviously have to reject the justified true belief account of knowledge, holding instead that knowledge is true belief which satisfies the chosen externalist condition, e.g., a result of a reliable process (and perhaps, further conditions as well). This makes it possible for such a view to retain internalist account of epistemic justification, though the centrality of that concept to epistemology would obviously be seriously diminished.
 Such an externalist account of knowledge can accommodate the commonsense conviction that animals, young children, and unsophisticated adults posses knowledge, though not the weaker conviction (if such a conviction does exists) that such individuals are epistemically justified in their beliefs. It is also at least less vulnerable to internalist counter-examples of the sort discussed, since the intuitions involved there pertain more clearly to justification than to knowledge. What is uncertain is what ultimate philosophical significance the resulting conception of knowledge is supposed to have. In particular, does it have any serious bearing on traditional epistemological problems and on the deepest and most troubling versions of scepticism, which seems in fact to be primarily concerned with justification, the an knowledge ?`
 A rather different use of the terms internalism and externalism has to do with the issue of how the content of beliefs and thoughts is determined : According to an internalist view of content, the content of such intention states depends only on the non-relational, internal properties of the individuals mind or grain, and not at all on his physical and social environment : While according to an externalist view, content is significantly affected by such external factors and suggests a view that appears of both internal and external elements are standardly classified as an external view.
 As with justification and knowledge, the traditional view of content has been strongly internalized in character. The main argument for externalism derives from the philosophy of language, more specifically from the various phenomena pertaining to natural kind terms, indexicals, etc. that motivate the views that have come to be known as direct reference theories. Such phenomena seem at least to show that the belief or thought content that can be properly attributed to a person is dependant on facts about his environment, e.g., whether he is on Earth or Twin Earth, what is fact pointing at, the classificatory criteria employed by expects in his social group, etc. - not just on what is going on internally in his mind or brain.
 An objection to externalist account of content is that they seem unable to do justice to our ability to know the content of our beliefs or thought from the inside, simply by reflection. If content is depend on external factors pertaining to the environment, then knowledge of content should depend on knowledge of these factors - which will not in general be available to the person whose belief or thought is in question.
 The adoption of an externalized account of mental content would seem to support an externalist account of justification, by way that if part or all of the content of a belief inaccessible to the believe, then both the justifying status of other beliefs in relation to that content and the status of that content justifying the beliefs will be similarly inaccessible, thus contravening the internalist requirement for justification. An internalist must insist that there are no justifiable relations of these sorts, that our internally associable content can either be justified or justly anything else : But such a response appears lame unless it is coupled with an attempt to show that the externalist account of content is mistaken.
 A great deal of philosophical effort has been lavished on the attempt to naturalize content, i.e. to explain in non-semantic, Non-intentional terms what it is for something to be represental (have content) and what it is for something to have some particular content rather than some other. There appear to be only four types of theory that have been proposed: Theories that ground representation in (1) similarity, (2) conversance, (3) functional role, (4) teleology.
 Similarly, theories hold that 'r' represents 'x' in virtue of being similar to 'x'. This has seemed hopeless to most as a theory of mental representation  because it appears to require that things in the brain must share properties with the things they represent: To represent a cat as furry appears to require something furry in the brain. Perhaps, a notion of similarity that is naturalistic and does not involve property sharing can be worked out, but it is not obvious how.
 Covariance theories hold that 'r's' represent 'x' is grounded in the fact that 'r's', occasion canaries with that of 'x'. This is most compelling he n one thinks about detection systems, the firing a neural structures in the visual system is said to represent vertical orientations, if its firing varies with the occurrence of vertical lines in the visual field of perceptivity.
 Functional role theories hold that 'r's' represent 'x' is grounded in the functional role 'r' has in the representing system, i.e., on the relations imposed by specific cognitive processes imposed by specific cognitive processes between 'r' and other representations in the system's repertoire. Functional role theories take their cue from such common-sense ideas as that people cannot believer that cats are furry if they did not know that cats are animals or that fur is like hair.
 Teleological theories hold that 'r' represent 'x' if it is 'r's' function to indicate, i.e., covary with 'x'. Teleological theories differ depending on the theory of functions they import. Perhaps the most important distinction is that between historical theories of functions. Historical theories individuated functional states (hence contents) in a way that is sensitive to the historical development of the state, i.e., to factors such as the way the state was 'learned', or the way it evolved. An historical theory might hold that the function of 'r' is to indicate 'x' only if the capacity to token 'r' was developed (selected, learned) because it indicates 'x'. Thus, a state physically indistinguishable from 'r's' historical origins would not represent 'x' according to historical theories.
 Theories of representational content may be classified according to whether they are atomistic or holistic and according to whether they are externalistic or internalistic, whereby, emphasizing the priority of a whole over its parts. Furthermore, in the philosophy of language, this becomes the claim that the meaning of an individual word or sentence can only be understood in terms of its relation to an indefinitely larger body of language, such as a whole theory, or even a whole language or form of life. In the philosophy of mind a mental state similarly may be identified only in terms of its relations with others. Moderate holism may allow the other things besides these relationships also count; extreme holism would hold that a network of relationships is all that we have. A holistic view of science holds that experience only confirms or disconfirms large bodies of doctrine, impinging at the edges, and leaving some leeway over the adjustment that it requires.
 Once, again, in the philosophy of mind and language, the view that what is thought, or said, or experienced, is essentially dependent on aspects of the world external to the mind of the subject. The view goes beyond holding that such mental states are typically caused by external factors, to insist that they could not have existed as they now do without the subject being embedded in an external world of a certain kind. It is these external relations that make up the essence or identify of the mental state. Externalism is thus opposed to the Cartesian separation of the mental from the physical, since that holds that the mental could in principle exist as it does even if there were no external world at all. Various external factors have been advanced as ones on which mental content depends, including the usage of experts, the linguistic, norms of the community. And the general causal relationships of the subject. In the theory of knowledge, externalism is the view that a person might know something by being suitably situated with respect to it, without that relationship being in any sense within his purview. The person might, for example, be very reliable in some respect without believing that he is. The view allows that you can know without being justified in believing that you know.
 However, atomistic theories take a representation's content to be something that can be specified independent entity of that representation' s relations to other representations. What the American philosopher of mind, Jerry Alan Fodor (1935-) calls the crude causal theory, for example, takes a representation to be a |cow| - a menial representation with the same content as the word 'cow' - if its tokens are caused by instantiations of the property of being-a-cow, and this is a condition that places no explicit constraints on how |cow|'s must or might relate to other representations. Holistic theories contrasted with atomistic theories in taking the relations a representation bears to others to be essential to its content. According to functional role theories, a representation is a |cow| if it behaves like a |cow| should behave in inference.
 Internalist theories take the content of a representation to be a matter determined by factors internal to the system that uses it. Thus, what Block (1986) calls 'short-armed' functional role theories are internalist. Externalist theories take the content of a representation to be determined, in part at least, by factors external to the system that uses it. Covariance theories, as well as teleological theories that invoke an historical theory of functions, take content to be determined by 'external' factors. Crossing the atomist-holistic distinction with the internalist-externalist distinction.
 Externalist theories (sometimes called non-individualistic theories) have the consequence that molecule for molecule are coincide with the identical cognitive systems might yet harbour representations with different contents. This has given rise to a controversy concerning 'narrow' content. If we assume some form of externalist theory is correct, then content is, in the first instance 'wide' content, i.e., determined in part by factors external to the representing system. On the other hand, it seems clear that, on plausible assumptions about how to individuate psychological capacities, internally equivalent systems must have the same psychological capacities. Hence, it would appear that wide content cannot be relevant to characterizing psychological equivalence. Since cognitive science generally assumes that content is relevant to characterizing psychological equivalence, philosophers attracted to externalist theories of content have sometimes attempted to introduce 'narrow' content, i.e., an aspect or kind of content that is equivalent internally equivalent systems. The simplest such theory is Fodor's idea (1987) that narrow content is a function from contents (i.e., from whatever the external factors are) to wide contents.
 All the same, what a person expresses by a sentence is often a function of the environment in which he or she is placed. For example, the disease I refer to by the term like 'arthritis', or the kind of tree I refer to as a 'Maple' will be defined by criteria of which I know next to nothing. This raises the possibility of imagining two persons in rather different environments, but in which everything appears the same to each of them. The wide content of their thoughts and sayings will be different if the situation surrounding them is appropriately different: 'situation' may include the actual objects they perceive or the chemical or physical kinds of object in the world they inhabit, or the history of their words, or the decisions of authorities on what counts as an example, of one of the terms they use. The narrow content is that part of their thought which remains identical, through their identity of the way things appear, regardless of these differences of surroundings. Partisans of wide content may doubt whether any content in this sense narrow, partisans of narrow content believer that it is the fundamental notion, with wide content being explicable in terms of narrow content plus context.
 Even so, the distinction between facts and values has outgrown its name: it applies not only to matters of fact vs, matters of value, but also to statements that something is, vs. statements that something ought to be. Roughly, factual statements - 'is statements' in the relevant sense - represent some state of affairs as obtaining, whereas normative statements - evaluative, and deontic ones - attribute goodness to something, or ascribe, to an agent, an obligation to act. Neither distinction is merely linguistic. Specifying a book's monetary value is making a factual statement, though it attributes a kind of value. 'That is a good book' expresses a value judgement though the term 'value' is absent (nor would 'valuable' be synonymous with 'good'). Similarly, 'we are morally obligated to fight' superficially expresses a statement, and 'By all indications it ought to rain' makes a kind of ought-claim; but the former is an ought-statement, the latter an (epistemic) is-statement.
 Theoretical difficulties also beset the distinction. Some have absorbed values into facts holding that all value is instrumental, roughly, to have value is to contribute - in a factual analysable way - to something further which is (say) deemed desirable. Others have suffused facts with values, arguing that facts (and observations) are 'theory-impregnated' and contending that values are inescapable to theoretical choice. But while some philosophers doubt that fact/value distinctions can be sustained, there persists a sense of a deep difference between evaluating, and attributing an obligation and, on the other hand, saying how the world is.
 Fact/value distinctions, may be defended by appeal to the notion of intrinsic value, value a thing has in itself and thus independently of its consequences. Roughly, a value statement (proper) is an ascription of intrinsic value, one to the effect that a thing is to some degree good in itself. This leaves open whether ought-statements are implicitly value statements, but even if they imply that something has intrinsic value - e.g., moral value - they can be independently characterized, say by appeal to rules that provide (justifying) reasons for action. One might also ground the fact value distinction in the attributional (or even motivational) component apparently implied by the making of valuational or deontic judgements: Thus, 'it is a good book, but that is no reason for a positive attribute towards it' and 'you ought to do it, but there is no reason to' seem inadmissible, whereas, substituting, 'an expensive book' and 'you will do it' yields permissible judgements. One might also argue that factual judgements are the kind which are in principle appraisable scientifically, and thereby anchor the distinction on the factual side. This ligne is plausible, but there is controversy over whether scientific procedures are 'value-free' in the required way.
 Philosophers differ regarding the sense, if any, in which epistemology is normative (roughly, valuational). But what precisely is at stake in this controversy is no clearly than the problematic fact/value distinction itself. Must epistemologists as such make judgements of value or epistemic responsibility? If epistemology is naturalizable, then even epistemic principles simply articulate under what conditions - say, appropriate perceptual stimulations - a belief is justified, or constitutes knowledge. Its standards of justification, then would be like standards of, e.g., resilience for bridges. It is not obvious, however, that there appropriate standards can be established without independent judgements that, say, a certain kind of evidence is good enough for justified belief (or knowledge). The most plausible view may be that justification is like intrinsic goodness, though it supervenes on natural properties, it cannot be analysed wholly in factual statements.
 Thus far, belief has been depicted as being all-or-nothing, however, as a resulting causality for which we have grounds for thinking it true, and, all the same, its acceptance is governed by epistemic norms, and, least of mention, it is partially subject to voluntary control and has functional affinities to belief. Still, the notion of acceptance, like that of degrees of belief, merely extends the standard picture, and does not replace it.
 Traditionally, belief has been of epistemological interest in its propositional guise: 'S' believes that 'p', where 'p' is a reposition towards which an agent, 'S' exhibits an attitude of acceptance. Not all belief is of this sort. If I trust you to say, I believer you. And someone may believer in Mr. Radek, or in a free-market economy, or in God. It is sometimes supposed that all belief is 'reducible' to propositional belief, belief-that. Thus, my believing you might be thought a matter of my believing, is,  perhaps, that what you say is true, and your belief in free markets or God, is a matter of your believing that free-market economies are desirable or that God exists.
 Some philosophers have followed St., Thomas Aquinas (1225-74), in supposing that to believer in God is simply to believer that certain truths hold while others argue that belief-in is a distinctive attitude, on that includes essentially an element of trust. More commonly, belief-in has been taken to involve a combination of propositional belief together with some further attitude.
 The moral philosopher Richard Price (1723-91) defends the claim that there are different sorts of belief-in, some, but not all reducible to beliefs-that. If you believer in God, you believer that God exists, that God is good, you believer that God is good, etc. But according to Price, your belief involves, in addition, a certain complex pro-attitude toward its object. Even so, belief-in outruns the evidence for the corresponding belief-that. Does this diminish its rationality? If belief-in presupposes believes-that, it might be thought that the evidential standards for the former must be, at least, as high as standards for the latter. And any additional pro-attitude might be thought to require a further layer of justification not required for cases of belief-that.
 Belief-in may be, in general, less susceptible to alternations in the face of unfavourable evidence than belief-that. A believe who encounters evidence against God's existence may remain unshaken in his belief, in part because the evidence does not bear on his pro-attitude. So long as this is united with his belief that God exists, the reasonably so in a way that an ordinary propositional belief that would not.
 Some philosophers think that the category of knowing for which true. Justified believing (accepting) is a requirement constituting only a species of Propositional knowledge, construed as an even broader category. They have proposed various examples of 'PK' that do not satisfy the belief and/ort justification conditions of the tripartite analysis. Such cases are often recognized by analyses of Propositional knowledge in terms of powers, capacities, or abilities. For instance, Alan R. White (1982) treats 'PK' as merely the ability to provide a correct answer to a possible question, however, White may be equating 'producing' knowledge in the sense of producing 'the correct answer to a possible question' with 'displaying' knowledge in the sense of manifesting knowledge. (White, 1982). The latter can be done even by very young children and some non-human animals independently of their being asked questions, understanding questions, or recognizing answers to questions. Indeed, an example that has been proposed as an instance of knowing that 'h' without believing or accepting that 'h' can be modified so as to illustrate this point. Two examples concern an imaginary person who has no special training or information about horses or racing, but who in an experiment persistently and correctly picks the winners of upcoming horseraces. If the example is modified so that the hypothetical 'seer' never picks winners but only muses over whether those horses wight win, or only reports those horses winning, this behaviour should be as much of a candidate for the person's manifesting knowledge that the horse in question will win as would be the behaviour of picking it as a winner.
 These considerations expose limitations in Edward Craig's analysis (1990) of the concept of knowing of a person's being a satisfactory information in relation to an inquirer who wants to find out whether or not 'h'. Craig realizes that counterexamples to his analysis appear to be constituted by Knower who is too recalcitrant to inform the inquirer, or to incapacitate to inform, or too discredited to be worth considering (as with the boy who cried 'Wolf'). Craig admits that this might make preferably some alternative view of knowledge as a different state that helps to explain the presence of the state of being a suitable informant when the latter does obtain. Such an alternate, which offers a recursive definition that concerns one's having the power to proceed in a way representing the state of affairs, causally involved in one's proceeding in this way. When combined with a suitable analysis of representing, this theory of propositional knowledge can be unified with a structurally similar analysis of knowing how to do something.
 Knowledge and belief, according to most epistemologists, knowledge entails belief, so that I cannot know that such and such is the case unless I believe that such and such am the case. Others think this entailment thesis can be rendered more accurately if we substitute for belief some closely related attitude. For instance, several philosophers would prefer to say that knowledge entail psychological certainties (Prichard, 1950 and Ayer, 1956) or conviction (Lehrer, 1974) or acceptance (Lehrer, 1989). Nonetheless, there are arguments against all versions of the thesis that knowledge requires having a belief-like attitude toward the known. These arguments are given by philosophers who think that knowledge and belief (or a facsimile) are mutually incompatible (the incomparability thesis), or by ones who say that knowledge does not entail belief, or vice versa, so that each may exist without the other, but the two may also coexist (the separability thesis).
 The incompatibility thesis is sometimes traced to Plato 429-347 Bc. , In view of his claim that knowledge is infallible while belief or opinion is fallible ('Republic' 476-9). But this claim would not support the thesis. Belief might be a component of an infallible form of knowledge in spite of the fallibility of belief. Perhaps, knowledge involves some factor that compensates for the fallibility of belief.
 A. Duncan-Jones (1939: Also Vendler, 1978) cites linguistic evidence to back up the incompatibility thesis. He notes that people often say 'I do not believe she is guilty. I know she is' and the like, which suggest that belief rule out knowledge. However, as Lehrer (1974) indicates, the above exclamation is only a more emphatic way of saying 'I do not just believe she is guilty, I know she is' where 'just' makes it especially clear that the speaker is signalling that she has something more salient than mere belief, not that she has something inconsistent with belief, namely knowledge. Compare: 'You do not hurt him, you killed him'.
 H.A. Prichard (1966) offers a defence of the incompatibility thesis that hinges on the equation of knowledge with certainty (both infallibility and psychological certitude) and the assumption that when we believe in the truth of a claim we are not certain about its truth. Given that belief always involves uncertainty while knowledge never dies, believing something rules out the possibility of knowing it. Unfortunately, however, Prichard gives 'us' no goods reason to grant that states of belief are never ones involving confidence. Conscious beliefs clearly involve some level of confidence, to suggest that we cease to believe things about which we are completely confident is bizarre.
 A.D. Woozley (1953) defends a version of the separability thesis. Woozley's version, which deals with psychological certainty rather than belief per se, is that knowledge can exist in the absence of confidence about the item known, although might also be accompanied by confidence as well. Woozley remarks that the test of whether I know something is 'what I can do, where what I can do may include answering questions'. On the basis of this remark he suggests that even when people are unsure of the truth of a claim, they might know that the claim is true. We unhesitatingly attribute knowledge to people who give correct responses on examinations even if those people show no confidence in their answers. Woozley acknowledges, however, that it would be odd for those who lack confidence to claim knowledge. It would be peculiar to say, 'I am unsure my answer is true: Still, I know it is correct'. But this tension Woozley explains using a distinction between conditions under which we are justified in making a claim (such as a claim to know something), and conditions under which the claim we make are true. While 'I know such and such' might be true even if I am unsure whether such and such holds, nonetheless it would be inappropriate for me to claim that I know that such and such unless I were sure of the truth of my claim.
 Colin Radford (1966) extends Woozley's defence of the separability thesis. In Radford's view, not only is knowledge compatible with the lack of certainty, it is also compatible with a complete lack of belief. He argues by example. In one example, Jean has forgotten that he learned some English history year's priori and yet he is able to give several correct responses to questions such as 'When did the Battle of Hastings occur'? Since he forgot that he took history, he considers the correct response to be no more than guesses. Thus, when he says that the Battle of Hastings took place in 1066 he would deny having the belief that the Battle of Hastings took place in 1066. A disposition he would deny being responsible (or having the right to be convincing) that 1066 was the correct date. Radford would nonetheless insist that Jean know when the Battle occurred, since clearly be remembering the correct date. Radford admits that it would be inappropriate for Jean to say that he knew when the Battle of Hastings occurred, but, like Woozley he attributes the impropriety to a fact about when it is and is not appropriate to claim knowledge. When we claim knowledge, we ought, at least to believe that we have the knowledge we claim, or else our behaviour is 'intentionally misleading'.
 Those that agree with Radford's defence of the separability thesis will probably think of belief as an inner state that can be detected through introspection. That Jean lack's beliefs about English history are plausible on this Cartesian picture since Jean does not find himself with any beliefs about English history when we seek them out. One might criticize Radford, however, by rejecting that Cartesian view of belief. One could argue that some beliefs are thoroughly unconscious, for example. Or one could adopt a behaviourist conception of belief, such as Alexander Bain's (1859), according to which having beliefs is a matter of the way people are disposed to behave (and has not Radford already adopted a behaviourist conception of knowledge?) Since Jean gives the correct response when queried, a form of verbal behaviour, a behaviourist would be tempted to credit him with the belief that the Battle of Hastings occurred in 1066.
 D.M. Armstrong (1873) takes a different tack against Radford. Jean does know that the Battle of Hastings took place in 1066. Armstrong will grant Radfod that point, in fact, Armstrong suggests that Jean believe that 1066 is not the date the Battle of Hastings occurred, for Armstrong equates the belief that such and such is just possible but no more than just possible with the belief that such and such is not the case. However, Armstrong insists, Jean also believes that the Battle did occur in 1066. After all, had Jean been mistaught that the Battle occurred in 1066, and subsequently 'guessed' that it took place in 1066, we would surely describe the situation as one in which Jean's false belief about the Battle became unconscious over time but persisted of a memory trace that was causally responsible for his guess. Out of consistency, we must describe Radford's original case as one that Jean's true belief became unconscious but persisted long enough to cause his guess. Thus, while Jean consciously believes that the Battle did not occur in 1066, unconsciously he does believe it occurred in 1066. So after all, Radford does not have a counterexample to the claim that knowledge entails belief.
 Armstrong's response to Radford was to reject Radford's claim that the examinee lacked the relevant belief about English history. Another response is to argue that the examinee lacks the knowledge Radford attributes to him (cf. Sorenson, 1982). If Armstrong is correct in suggesting that Jean believes both that 1066 is and that it is not the date of the Battle of Hastings, one might deny Jean knowledge on the grounds that people who believe the denial of what they believe cannot be said t know the truth of their belief. Another strategy might be to compare the examine case with examples of ignorance given in recent attacks on externalist accounts of knowledge (needless to say.  Externalists themselves would tend not to favour this strategy). Consider the following case developed by BonJour (1985): For no apparent reason, Samantha believes that she is clairvoyant. Again, for no apparent reason, she one day comes to believe that the President is in New York City, even though she has every reason to believe that the President is in Washington, D.C. In fact, Samantha is a completely reliable clairvoyant, and she has arrived at her belief about the whereabouts of the President thorough the power of her clairvoyance. Yet surely Samantha's belief is completely irrational. She is not justified in thinking what she does. If so, then she does not know where the President is. But Radford's examinee is unconventional. Even if Jean lacks the belief that Radford denies him, Radford does not have an example of knowledge that is unattended with belief. Suppose that Jean's memory had been sufficiently powerful to produce the relevant belief. As Radford says, in having every reason to suppose that his response is mere guesswork, and he has every reason to consider his belief false. His belief would be an irrational one, and hence one about whose truth Jean would be ignorant.
 Least has been of mention to an approaching view from which 'perception' basis upon itself as a fundamental philosophical topic both for its central place in ant theory of knowledge, and its central place un any theory of consciousness. Philosophy in this area is constrained by a number of properties that we believe to hold of perception, (1) It gives 'us' knowledge of the world around 'us'. (2) We are conscious of that world by being aware of 'sensible qualities': Colour, sounds, tastes, smells, felt warmth, and the shapes and positions of objects in the environment. (3) Such consciousness is affected through highly complex information channels, such as the output of the three different types of colour-sensitive cells in the eye, or the channels in the ear for interpreting pulses of air pressure as frequencies of sound. (4) There ensues even more complex neurophysiological coding of that information, and eventually higher-order brain functions bring it about that we interpreted the information so received. (Much of this complexity has been revealed by the difficulties of writing programs enabling computers to recognize quite simple aspects of the visual scene.) The problem is to avoid thinking of here being a central, ghostly, conscious self, fed information in the same way that a screen if fed information by a remote television camera. Once such a model is in place, experience will seem like a veil getting between 'us' and the world, and the direct objects of perception will seem to be private items in an inner theatre or sensorium. The difficulty of avoiding this model is epically cute when we considered the secondary qualities of colour, sound, tactile feelings and taste, which can easily seem to have a purely private existence inside the perceiver, like sensation of pain. Calling such supposed items names like 'sense-data' or 'percepts' exacerbate the tendency, but once the model is in place, the first property, that perception gives 'us' knowledge of the world and its surrounding surfaces, is quickly threatened, for there will now seem little connection between these items in immediate experience and any independent reality. Reactions to this problem include 'scepticism' and 'idealism'.
 A more hopeful approach is to claim that the complexities of (3) and (4) explain how we can have direct acquaintance of the world, than suggesting that the acquaintance we do have been at best indirect. It is pointed out that perceptions are not like sensation, precisely because they have a content, or outer-directed nature. To have a perception is to be aware of the world for being such-and-such a way, than to enjoy a mere modification of sensation. But such direct realism has to be sustained in the face of the evident personal (neurophysiological and other) factors determining how we perceive. One approach is to ask why it is useful to be conscious of what we perceive, when other aspects of our functioning work with information determining responses without any conscious awareness or intervention. A solution to this problem would offer the hope of making consciousness part of the natural world, than a strange optional extra.
 Furthering, perceptual knowledge is knowledge acquired by or through the senses and includes most of what we know. We cross intersections when we see the light turn green, head for the kitchen when we smell the roast burning, squeeze the fruit to determine its ripeness, and climb out of bed when we hear the alarm ring. In each case we come to know something-that the light has turned green, that the roast is burning, that the melon is overripely, and that it is time to get up-by some sensory means. Seeing that the light has turned green is learning something-that, the light has turned green-by use of the eyes. Feeling that the melon is overripe is coming to know a fact-that the melon is overripe-by one's sense to touch. In each case the resulting knowledge is somehow based on, derived from or grounded in the sort of experience that characterizes the sense modality in question.
 Much of our perceptual knowledge is indirect, dependent or derived. By this I mean that the facts we describe ourselves as learning, as coming to know, by perceptual means are pieces of knowledge that depend on our coming to know something else, some other fact, in a more direct way. We see, by the gauge, that we need gas, see, by the newspapers, that our team has lost again, see, by her expression, that she is nervous. This derived or dependent sort of knowledge is particularly prevalent in the cases of vision, but it occurs, to a lesser degree, in every sense modality. We install bells and other noise-makers so that we calm for example, hear (by the bell) that someone is at the door and (by the alarm) that its time to get up. When we obtain knowledge in this way, it is clear that unless one sees-hence, comes to know something about the gauge (that it says) and (hence, know) that one is described as coming to know by perceptual means. If one cannot hear that the bell is ringing, one cannot-in at least in this way-hear that one's visitors have arrived. In such cases one sees (hears, smells, etc.) that 'a' is 'F', coming to know thereby that 'a' is 'F', by seeing (hearing, etc.) that some other condition, 'b's' being 'G', obtains when this occurs, the knowledge (that 'a' is 'F') is derived from, or dependent on, the more basic perceptual knowledge that 'b' is 'G'.
 Perhaps as a better strategy is to tie an account save that part that evidence could justify explanation for it is its truth alone. Since, at least the times of Aristotle philosophers of explanatory knowledge have emphasized of its importance that, in its simplest therms, we want to know not only what is the composite peculiarities and particular points of issue but also why it is. This consideration suggests that we define an explanation as an answer to a why-question. Such a definition would, however, be too broad, because some why-questions are requests for consolation (Why did my son have to die?) Or moral justification (Why should women not be paid the same as men for the same work?) It would also be too narrow because some explanations are responses to how-questions (How does radar work?) Or how-possibility-questions (How is it possible for cats always to land their feet?)
 In its overall sense, 'to explain' means to make clear, to make plain, or to provide understanding. Definitions of this sort are philosophically unhelpful, for the terms used in the deficient are no less problematic than the term to be defined. Moreover, since a wide variety of things require explanation, and since many different types of explanation exist, as more complex explanation is required. To facilitate the requirement leaves, least of mention, for us to consider by introduction a bit of technical terminology. The term 'explanation' is used to refer to that which is to be explained: The term 'explanans' refer to that which does the explaining, the explanans and the explanation taken together constitute the explanation.
 One common type of explanation occurs when deliberate human actions are explained in terms of conscious purposes. 'Why did you go to the pharmacy yesterday?' 'Because I had a headache and needed to get some aspirin.' It is tacitly assumed that aspirin is an appropriate medication for headaches and that going to the pharmacy would be an efficient way of getting some. Such explanations are, of course, teleological, referring, ss they do, to goals. The explanans are not the realisation of a future goal - if the pharmacy happened to be closed for stocktaking the aspirin would have been obtained there, bu t that would not invalidate the explanation. Some philosophers would say that the antecedent desire to achieve the end is what doers the explaining: Others might say that the explaining is done by the nature of the goal and the fact that the action promoted the chances of realizing it. (Taylor, 1964). In that it should not be automatically being assumed that such explanations are causal. Philosophers differ considerably on whether these explanations are to be framed in terms of cause or reason, but the distinction cannot be used to show that the relation between reasons and the actions they justify is in no way causal, and there are many differing analyses of such concepts as intention and agency. Expanding the domain beyond consciousness, Freud maintained, in addition, that much human behaviour can be explained in terms of unconscious and conscious wishes. Those Freudian explanations should probably be construed as basically causal.
 Problems arise when teleological explanations are offered in other context. The behaviour of non-human animals is often explained in terms of purpose, e.g., the mouse ran to escape from the cat. In such cases the existence of conscious purpose seems dubious. The situation is still more problematic when a supr-empirical purpose in invoked, e.g., the explanations of living species in terms of God's purpose, or the vitalistic explanations of biological phenomena in terms of a entelechy or vital principle. In recent years an 'anthropic principle' has received attention in cosmology (Barrow and Tipler, 1986). All such explanations have been condemned by many philosophers an anthropomorphic.
 Nevertheless, philosophers and scientists often maintain that functional explanations play an important an legitimate role in various sciences such as, evolutionary biology, anthropology and sociology. For example, of the peppered moth in Liverpool, the change in colour from the light phase to the dark phase and back again to the light phase provided adaption to a changing environment and fulfilled the function of reducing predation on the spacies. In the study of primitive soviets anthropologists have insisted that various rituals the (rain dance) which may be inefficacious in braining about their manifest goals (producing rain), actually cohesion at a period of stress (often a drought). Philosophers who admit teleological and/or functional explanations in common sense and science oftentimes take pans to argue that such explanations can be annualized entirely in terms of efficient causes, thereby escaping the charge of anthropomorphism (Wright, 1976): Again, however, not all philosophers agree.
 Causal theories of Propositional knowledge differ over whether they deviate from the tripartite analysis by dropping the requirements that one's believing (accepting) that 'h' be justified. The same variation occurs regarding reliability theories, which present the Knower as reliable concerning the issue of whether or not 'h', in the sense that some of one's cognitive or epistemic states, ?, are such that, given further characteristics of oneself-possibly including relations to factors external to one and which one may not be aware-it is nomologically necessary (or at least probable) that 'h'. In some versions, the reliability is required to be 'global' in as far as it must concern a nomologically (probabilistic-relationship) relationship that states of type ? to the acquisition of true beliefs about a wider range of issues than merely whether or not 'h'. There is also controversy about how to delineate the limits of what constitutes a type of relevant personal state or characteristic. (For example, in a case where Mr Notgot has not been shamming and one does know thereby that someone in the office owns a Ford, such as a way of forming beliefs about the properties of persons spatially close to one, or instead something narrower, such as a way of forming beliefs about Ford owners in offices partly upon the basis of their relevant testimony?)
 One important variety of reliability theory is a conclusive reason account, which includes a requirement that one's reasons for believing that 'h' be such that in one's circumstances, if h* were not to occur then, e.g., one would not have the reasons one does for believing that 'h', or, e.g., one would not believe that 'h'. Roughly, the latter is demanded by theories that treat a Knower as 'tracking the truth', theories that include the further demand that is roughly, if it were the case, that 'h', then one would believe that 'h'. A version of the tracking theory has been defended by Robert Nozick (1981), who adds that if what he calls a 'method' has been used to arrive at the belief that 'h', then the antecedent clauses of the two conditionals that characterize tracking will need to include the hypothesis that one would employ the very same method.
 But unless more conditions are added to Nozick's analysis, it will be too weak to explain why one lack's knowledge in a version of the last variant of the tricky Mr Notgot case described above, where we add the following details: (a) Mr Notgot's compulsion is not easily changed, (b) while in the office, Mr Notgot has no other easy trick of the relevant type to play on one, and finally for one's belief that 'h', not by reasoning through a false belief ut by basing belief that 'h', upon a true existential generalization of one's evidence.
 Nozick's analysis is in addition too strong to permit anyone ever to know that 'h': 'Some of my beliefs about beliefs might be otherwise, e.g., I might have rejected on of them'. If I know that 'h5' then satisfaction of the antecedent of one of Nozick's conditionals would involve its being false that 'h5', thereby thwarting satisfaction of the consequent's requirement that I not then believe that 'h5'. For the belief that 'h5' is itself one of my beliefs about beliefs (Shope, 1984).
 Some philosophers think that the category of knowing for which is true. Justified believing (accepting) is a requirement constituting only a species of Propositional knowledge, construed as an even broader category. They have proposed various examples of 'PK' that do not satisfy the belief and/ort justification conditions of the tripartite analysis. Such cases are often recognized by analyses of Propositional knowledge in terms of powers, capacities, or abilities. For instance, Alan R. White (1982) treats 'PK' as merely the ability to provide a correct answer to a possible question. White may be equating 'producing' knowledge in the sense of producing 'the correct answer to a possible question' with 'displaying' knowledge in the sense of manifesting knowledge. (White, 1982). The latter can be done even by very young children and some non-human animals independently of their being asked questions, understanding questions, or recognizing answers to questions. Indeed, an example that has been proposed as an instance of knowing that 'h' without believing or accepting that 'h' can be modified so as to illustrate this point. Two examples concerns an imaginary person who has no special training or information about horses or racing, but who in an experiment persistently and correctly picks the winners of upcoming horseraces. If the example is modified so that the hypothetical 'seer' never picks winners but only muses over whether those horses wight win, or only reports those horses winning, this behaviour should be as much of a candidate for the person's manifesting knowledge that the horse in question will win as would be the behaviour of picking it as a winner.
 These considerations expose limitations in Edward Craig's analysis (1990) of the concept of knowing of a person's being a satisfactory informants in relation to an inquirer who wants to find out whether or not 'h'. Craig realizes that counterexamples to his analysis appear to be constituted by Knower who are too recalcitrant to inform the inquirer, or too incapacitate to inform, or too discredited to be worth considering (as with the boy who cried 'Wolf'). Craig admits that this might make preferable some alternative view of knowledge as a different state that helps to explain the presence of the state of being a suitable informant when the latter does obtain. Such the alternate, which offers a recursive definition that concerns one's having the power to proceed in a way representing the state of affairs, causally involved in one's proceeding in this way. When combined with a suitable analysis of representing, this theory of propositional knowledge can be unified with a structurally similar analysis of knowing how to do something.
 Knowledge and belief, according to most epistemologists, knowledge entails belief, so that I cannot know that such and such is the case unless I believe that such and such is the case. Others think this entailment thesis can be rendered more accurately if we substitute for belief some closely related attitude. For instance, several philosophers would prefer to say that knowledge entail psychological certainties (Prichard, 1950 and Ayer, 1956) or conviction (Lehrer, 1974) or acceptance (Lehrer, 1989). Nonetheless, there are arguments against all versions of the thesis that knowledge requires having a belief-like attitude toward the known. These arguments are given by philosophers who think that knowledge and belief (or a facsimile) are mutually incompatible (the incomparability thesis), or by ones who say that knowledge does not entail belief, or vice versa, so that each may exist without the other, but the two may also coexist (the separability thesis).
 The incompatibility thesis is sometimes traced to Plato (429-347 Bc) in view of his claim that knowledge is infallible while belief or opinion is fallible ('Republic' 476-9). But this claim would not support the thesis. Belief might be a component of an infallible form of knowledge in spite of the fallibility of belief. Perhaps, knowledge involves some factor that compensates for the fallibility of belief.
 A. Duncan-Jones (1939: Also Vendler, 1978) cite linguistic evidence to back up the incompatibility thesis. He notes that people often say 'I do not believe she is guilty. I know she is' and the like, which suggest that belief rule out knowledge. However, as Lehrer (1974) indicates, the above exclamation is only a more emphatic way of saying 'I do not just believe she is guilty, I know she is' where 'just' makes it especially clear that the speaker is signalling that she has something more salient than mere belief, not that she has something inconsistent with belief, namely knowledge. Compare: 'You do not hurt him, you killed him.'
 H.A. Prichard (1966) offers a defence of the incompatibility thesis that hinges on the equation of knowledge with certainty (both infallibility and psychological certitude) and the assumption that when we believe in the truth of a claim we are not certain about its truth. Given that belief always involves uncertainty while knowledge never dies, believing something rules out the possibility of knowing it. Unfortunately, however, Prichard gives 'us' no goods reason to grant that states of belief are never ones involving confidence. Conscious beliefs clearly involve some level of confidence, to suggest that we cease to believe things about which we are completely confident is bizarre.
 A.D. Woozley (1953) defends a version of the separability thesis. Woozley's version, which deals with psychological certainty rather than belief per se, is that knowledge can exist in the absence of confidence about the item known, although might also be accompanied by confidence as well. Woozley remarks that the test of whether I know something is 'what I can do, where what I can do may include answering questions.' On the basis of this remark he suggests that even when people are unsure of the truth of a claim, they might know that the claim is true. We unhesitatingly attribute knowledge to people who give correct responses on examinations even if those people show no confidence in their answers. Woozley acknowledges, however, that it would be odd for those who lack confidence to claim knowledge. It would be peculiar to say, I am unsure that for whatever reason my answer is true: Still, I know it is correct But this tension Woozley explains using a distinction between conditions under which we are justified in making a claim (such as a claim to know something), and conditions under which the claim we make is true. While 'I know such and such' might be true even if I am unsure whether such and such holds, nonetheless it would be inappropriate for me to claim that I know that such and such unless I were sure of the truth of my claim.
 Colin Radford (1966) extends Woozley's defence of the separability thesis. In Radford's view, not only is knowledge compatible with the lack of certainty, it is also compatible with a complete lack of belief. He argues by example. In one example, Jean has forgotten that he learned some English history year's priori and yet he is able to give several correct responses to questions such as 'When did the Battle of Hastings occur?' Since he forgot that he took history, he considers the correct response to be no more than guesses. Thus, when he says that the Battle of Hastings took place in 1066 he would deny having the belief that the Battle of Hastings took place in 1066. A disposition he would deny being responsible (or having the right to be convincing) that 1066 was the correct date. Radford would none the less insist that Jean know when the Battle occurred, since clearly be remembering the correct date. Radford admits that it would be inappropriate for Jean to say that he knew when the Battle of Hastings occurred, but, like Woozley he attributes the impropriety to a fact about when it is and is not appropriate to claim knowledge. When we claim knowledge, we ought, at least to believe that we have the knowledge we claim, or else our behaviour is 'intentionally misleading'.
 Those that agree with Radford's defence of the separability thesis will probably think of belief as an inner state that can be detected through introspection. That Jean lack's beliefs about English history is plausible on this Cartesian picture since Jean does not find himself with any beliefs about English history when ne seek them out. One might criticize Radford, however, by rejecting that Cartesian view of belief. One could argue that some beliefs are thoroughly unconscious, for example. Or one could adopt a behaviourist conception of belief, such as Alexander Bain's (1859), according to which having beliefs is a matter of the way people are disposed to behave (and has not Radford already adopted a behaviourist conception of knowledge?) Since Jean gives the correct response when queried, a form of verbal behaviour, a behaviourist would be tempted to credit him with the belief that the Battle of Hastings occurred in 1066.
 D.M. Armstrong (1873) takes a different tack against Radford. Jean does know that the Battle of Hastings took place in 1066. Armstrong will grant Radfod that point, in fact, Armstrong suggests that Jean believe that 1066 is not the date the Battle of Hastings occurred, for Armstrong equates the belief that such and such is just possible but no more than just possible with the belief that such and such is not the case. However, Armstrong insists, Jean also believes that the Battle did occur in 1066. After all, had Jean been mistaught that the Battle occurred in 1066, and subsequently 'guessed' that it took place in 1066, we would surely describe the situation as one in which Jean's false belief about the Battle became unconscious over time but persisted of a memory trace that was causally responsible for his guess. Out of consistency, we must describe Radford's original case as one that Jean's true belief became unconscious but persisted long enough to cause his guess. Thus, while Jean consciously believes that the Battle did not occur in 1066, unconsciously he does believe it occurred in 1066. So after all, Radford does not have a counterexample to the claim that knowledge entails belief.
 Armstrong's response to Radford was to reject Radford's claim that the examinee lacked the relevant belief about English history. Another response is to argue that the examinee lacks the knowledge Radford attributes to him (cf. Sorenson, 1982). If Armstrong is correct in suggesting that Jean believes both that 1066 is and that it is not the date of the Battle of Hastings, one might deny Jean knowledge on the grounds that people who believe the denial of what they believe cannot be said t know the truth of their belief. Another strategy might be to compare the examine case with examples of ignorance given in recent attacks on externalist accounts of knowledge (needless to say.  Externalists themselves would tend not to favour this strategy). Consider the following case developed by BonJour (1985): For no apparent reason, Samantha believes that she is clairvoyant. Again, for no apparent reason, she one day comes to believe that the President is in New York City, even though she has every reason to believe that the President is in Washington, D.C. In fact, Samantha is a completely reliable clairvoyant, and she has arrived at her belief about the whereabouts of the President thorough the power of her clairvoyance. Yet surely Samanthas belief is completely irrational. She is not justified in thinking what she does. If so, then she does not know where the President is. But Radford's examinee is unconventional. Even if Jean lacks the belief that Radford denies him, Radford does not have an example of knowledge that is unattended with belief. Suppose that Jean's memory had been sufficiently powerful to produce the relevant belief. As Radford says, in having every reason to suppose that his response is mere guesswork, and he has every reason to consider his belief false. His belief would be an irrational one, and hence one about whose truth Jean would be ignorant.
 Least has been of mention to an approaching view from which 'perception' basis upon itself as a fundamental philosophical topic both for its central place in ant theory of knowledge, and its central place un any theory of consciousness. Philosophy in this area is constrained by a number of properties that we believe to hold of perception, (1) It gives 'us' knowledge of the world around 'us,' (2) We are conscious of that world by being aware of 'sensible qualities': Colour, sounds, tastes, smells, felt warmth, and the shapes and positions of objects in the environment. (3) Such consciousness is effected through highly complex information channels, such as the output of the three different types of colour-sensitive cells in the eye, or the channels in the ear for interpreting pulses of air pressure as frequencies of sound. (4) There ensues even more complex neurophysiological coding of that information, and eventually higher-order brain functions bring it about that we interpreted the information so received. (Much of this complexity has been revealed by the difficulties of writing programs enabling computers to recognize quite simple aspects of the visual scene.) The problem is to avoid thinking of here being a central, ghostly, conscious self, fed information in the same way that a screen if fed information by a remote television camera. Once such a model is in place, experience will seem like a veil getting between 'us' and the world, and the direct objects of perception will seem to be private items in an inner theatre or sensorium. The difficulty of avoiding this model is epically cute when we considered the secondary qualities of colour, sound, tactile feelings and taste, which can easily seem to have a purely private existence inside the perceiver, like sensation of pain. Calling such supposed items names like 'sense-data' or 'percepts' exacerbates the tendency, but once the model is in place, the first property, that perception gives 'us' knowledge of the world and its surrounding surfaces, is quickly threatened, for there will now seem little connection between these items in immediate experience and any independent reality. Reactions to this problem include 'scepticism' and 'idealism.'
 A more hopeful approach is to claim that the complexities of (3) and (4) explain how we can have direct acquaintance of the world, than suggesting that the acquaintance we do have been at best indirect. It is pointed out that perceptions are not like sensation, precisely because they have a content, or outer-directed nature. To have a perception is to be aware of the world for being such-and-such a way, than to enjoy a mere modification of sensation. But such direct realism has to be sustained in the face of the evident personal (neurophysiological and other) factors determining how we perceive. One approach is to ask why it is useful to be conscious of what we perceive, when other aspects of our functioning work with information determining responses without any conscious awareness or intervention. A solution to this problem would offer the hope of making consciousness part of the natural world, than a strange optional extra.
 Furthering, perceptual knowledge is knowledge acquired by or through the senses and includes most of what we know. We cross intersections when we see the light turn green, head for the kitchen when we smell the roast burning, squeeze the fruit to determine its ripeness, and climb out of bed when we hear the alarm ring. In each case we come to know something-that the light has turned green, that the roast is burning, that the melon is overripe, and that it is time to get up-by some sensory means. Seeing that the light has turned green is learning something-that, the light has turned green-by use of the eyes. Feeling that the melon is overripe is coming to know a fact-that the melon is overripe-by one's sense to touch. In each case the resulting knowledge is somehow based on, derived from or grounded in the sort of experience that characterizes the sense modality in question.
 Much as much of our perceptual knowledge is indirect, dependent or derived. By this I mean that the facts we describe ourselves as learning, as coming to know, by perceptual means are pieces of knowledge that depend on our coming to know something else, some other fact, in a more direct way. We see, by the gauge, that we need gas, see, by the newspapers, that our team has lost again, see, by her expression, that she is nervous. This derived or dependent sort of knowledge is particularly prevalent in the cases of vision, but it occurs, to a lesser degree, in every sense modality. We install bells and other noise-makers so that we calm for example, hear (by the bell) that someone is at the door and (by the alarm) that its time to get up. When we obtain knowledge in this way, it is clear that unless one sees-hence, comes to know something about the gauge (that it says) and (hence, know) that one is described as coming to know by perceptual means. If one cannot hear that the bell is ringing, one cannot-in at least in this way-hear that one's visitors have arrived. In such cases one sees (hears, smells, etc.) that 'a' is 'F', coming to know thereby that 'a' is 'F', by seeing (hearing, etc.) that some other condition, 'b's' being 'G', obtains when this occurs, the knowledge (that 'a' is 'F') is derived from, or dependent on, the more basic perceptual knowledge that 'b' is 'G'.
 And finally, the representational Theory of mind, (which goes back at least to Aristotle) takes as its starting point commonsense mental states, such as thoughts, beliefs, desires, perceptions and images. Such states are said to have 'intentionality' - they are about or refer to things, and may be evaluated with respect to properties like consistency, truth, appropriateness and accuracy. (For example, the thought that cousins are not related is inconsistent, the belief that Elvis is dead is true, the desire to eat the moon is inappropriate, a visual experience of a ripe strawberry as red is accurate, an image of George W. Bush with dreadlocks is inaccurate.)
 The Representational Theory of Mind, defines such intentional mental states as relations to mental representations, and explains the intentionality of the former in terms of the semantic properties of the latter. For example, to believe that Elvis is dead is to be appropriately related to a mental representation whose propositional content is that Elvis is dead. (The desire that Elvis be dead, the fear that he is dead, the regret that he is dead, etc., involve different relations to the same mental representation.) To perceive a strawberry is to have a sensory experience of some kind which is appropriately related to (e.g., caused by) the strawberry Representational theory of mind also understands mental processes such as thinking, reasoning and imagining as sequences of intentional mental states. For example, to imagine the moon rising over a mountain is to entertain a series of mental images of the moon (and a mountain). To infer a proposition q from the proposition's p and if 'p' then 'q' is (among other things) to have a sequence of thoughts of the form 'p', 'if p' then 'q', 'q'.
 Contemporary philosophers of mind have typically supposed (or at least hoped) that the mind can be naturalized -, i.e., that all mental facts have explanations in the terms of natural science. This assumption is shared within cognitive science, which attempts to provide accounts of mental states and processes in terms (ultimately) of features of the brain and central nervous system. In the course of doing so, the various sub-disciplines of cognitive science (including cognitive and computational psychology and cognitive and computational neuroscience) postulate a number of different kinds of structures and processes, many of which are not directly implicated by mental states and processes as commonsensical conceived. There remains, however, a shared commitment to the idea that mental states and processes are to be explained in terms of mental representations.
 In philosophy, recent debates about mental representation have centred around the existence of propositional attitudes (beliefs, desires, etc.) and the determination of their contents (how they come to be about what they are about), and the existence of phenomenal properties and their relation to the content of thought and perceptual experience. Within cognitive science itself, the philosophically relevant debates have been focussed on the computational architecture of the brain and central nervous system, and the compatibility of scientific and commonsense accounts of mentality.
 Intentional Realists such as Dretske (e.g., 1988) and Fodor (e.g., 1987) note that the generalizations we apply in everyday life in predicting and explaining each other's behaviour (often collectively referred to as 'folk psychology') are both remarkably successful and indispensable. What a person believes, doubts, desires, fears, etc. is a highly reliable indicator of what that person will do.  We have no other way of making sense of each other's behaviour than by ascribing such states and applying the relevant generalizations. We are thus committed to the basic truth of commonsense psychology and, hence, to the existence of the states its generalizations refer to. (Some realists, such as Fodor, also hold that commonsense psychology will be vindicated by cognitive science, given that propositional attitudes can be construed as computational relations to mental representations.)
 Intentional Eliminativists, such as Churchland, (perhaps) Dennett and (at one time) Stich argue that no such things as propositional attitudes (and their constituent representational states) are implicated by the successful explanation and prediction of our mental lives and behaviour. Churchland denies that the generalizations of commonsense propositional-attitude psychology are true. He (1981) argues that folk psychology is a theory of the mind with a long history of failure and decline, and that it resists incorporation into the framework of modern scientific theories (including cognitive psychology). As such, it is comparable to alchemy and phlogiston theory, and ought to suffer a comparable fate. Commonsense psychology is false, and the states (and representations) it postulates simply don't exist. (It should be noted that Churchland is not an eliminativist about mental representation tout court.
 Dennett (1987) grants that the generalizations of commonsense psychology are true and indispensable, but denies that this is sufficient reason to believe in the entities they appear to refer to. He argues that to give an intentional explanation of a system's behaviour is merely to adopt the 'intentional stance' toward it. If the strategy of assigning contentful states to a system and predicting and explaining its behaviour (on the assumption that it is rational -, i.e., that it behaves as it should, given the propositional attitudes it should have in its environment) is successful, then the system is intentional, and the propositional-attitude generalizations we apply to it are true. But there is nothing more to having a propositional attitude than this.
 Though he has been taken to be thus claiming that intentional explanations should be construed instrumentally, Dennett (1991) insists that he is a 'moderate' realist about propositional attitudes, since he believes that the patterns in the behaviour and behavioural dispositions of a system on the basis of which we (truly) attribute intentional states to it are objectively real. In the event that there are two or more explanatorily adequate but substantially different systems of intentional ascriptions to an individual, however, Dennett claims there is no fact of the matter about what the system believes (1987, 1991). This does suggest an irrealism at least with respect to the sorts of things Fodor and Dretske take beliefs to be; though it is not the view that there is simply nothing in the world that makes intentional explanations true.
 (Davidson 1973, 1974 and Lewis 1974 also defend the view that what it is to have a propositional attitude is just to be interpretable in a particular way. It is, however, not entirely clear whether they intend their views to imply irrealism about propositional attitudes.). Stich (1983) argues that cognitive psychology does not (or, in any case, should not) taxonomize mental states by their semantic properties at all, since attribution of psychological states by content is sensitive to factors that render it problematic in the context of a scientific psychology. Cognitive psychology seeks causal explanations of behaviour and cognition, and the causal powers of a mental state are determined by its intrinsic 'structural' or 'syntactic' properties. The semantic properties of a mental state, however, are determined by its extrinsic properties -, e.g., its history, environmental or intra-mental relations. Hence, such properties cannot figure in causal-scientific explanations of behaviour. (Fodor 1994 and Dretske 1988 are realist attempts to come to grips with some of these problems.) Stich proposes a syntactic theory of the mind, on which the semantic properties of mental states play no explanatory role.
 It is a traditional assumption among realists about mental representations that representational states come in two basic varieties (Boghossian 1995). There are those, such as thoughts, which are composed of concepts and have no phenomenal ('what-it's-like') features ('qualia'), and those, such as sensory experiences, which have phenomenal features but no conceptual constituents. (Non-conceptual content is usually defined as a kind of content that states of a creature lacking concepts but, nonetheless enjoy. On this taxonomy, mental states can represent either in a way analogous to expressions of natural languages or in a way analogous to drawings, paintings, maps or photographs. (Perceptual states such as seeing that something is blue, are sometimes thought of as hybrid states, consisting of, for example, a Non-conceptual sensory experience and a thought, or some more integrated compound of sensory and conceptual components.)
 Some historical discussions of the representational properties of mind (e.g., Aristotle 1984, Locke 1689/1975, Hume 1739/1978) seem to assume that Non-conceptual representations - percepts ('impressions'), images ('ideas') and the like - are the only kinds of mental representations, and that the mind represents the world in virtue of being in states that resemble things in it. On such a view, all representational states have their content in virtue of their phenomenal features. Powerful arguments, however, focussing on the lack of generality (Berkeley 1975), ambiguity (Wittgenstein 1953) and non-compositionality (Fodor 1981) of sensory and imaginistic representations, as well as their unsuitability to function as logical (Frége 1918/1997, Geach 1957) or mathematical (Frége 1884/1953) concepts, and the symmetry of resemblance (Goodman 1976), convinced philosophers that no theory of mind can get by with only Non-conceptual representations construed in this way.
 Contemporary disagreement over Non-conceptual representation concerns the existence and nature of phenomenal properties and the role they play in determining the content of sensory experience. Dennett (1988), for example, denies that there are such things as qualia at all; while Brandom (2002), McDowell (1994), Rey (1991) and Sellars (1956) deny that they are needed to explain the content of sensory experience. Among those who accept that experiences have phenomenal content, some (Dretske, Lycan, Tye) argue that it is reducible to a kind of intentional content, while others (Block, Loar, Peacocke) argue that it is irreducible.
 The representationalist thesis is often formulated as the claim that phenomenal properties are representational or intentional. However, this formulation is ambiguous between a reductive and a non-deductive claim (though the term 'representationalism' is most often used for the reductive claim). On one hand, it could mean that the phenomenal content of an experience is a kind of intentional content (the properties it represents). On the other, it could mean that the (irreducible) phenomenal properties of an experience determine an intentional content. Representationalists such as Dretske, Lycan and Tye would assent to the former claim, whereas phenomenalists such as Block, Chalmers, Loar and Peacocke would assent to the latter. (Among phenomenalists, there is further disagreement about whether qualia are intrinsically representational (Loar) or not (Block, Peacocke).
 Most (reductive) representationalists are motivated by the conviction that one or another naturalistic explanation of intentionality is, in broad outline, correct, and by the desire to complete the naturalization of the mental by applying such theories to the problem of phenomenality. (Needless to say, most phenomenalists (Chalmers is the major exception) are just as eager to naturalize the phenomenal - though not in the same way.)
 The main argument for representationalism appeals to the transparency of experience. The properties that characterize what it's like to have a perceptual experience are presented in experience as properties of objects perceived: in attending to an experience, one seems to 'see through it' to the objects and properties it is experiences of. They are not presented as properties of the experience itself. If nonetheless they were properties of the experience, perception would be massively deceptive. But perception is not massively deceptive. According to the representationalist, the phenomenal character of an experience is due to its representing objective, non-experiential properties. (In veridical perception, these properties are locally instantiated; in illusion and hallucination, they are not.) On this view, introspection is indirect perception: one comes to know what phenomenal features one's experience has by coming to know what objective features it represents.
 In order to account for the intuitive differences between conceptual and sensory representations, representationalists appeal to their structural or functional differences. Dretske (1995), for example, distinguishes experiences and thoughts on the basis of the origin and nature of their functions: an experience of a property 'P' is a state of a system whose evolved function is to indicate the presence of 'P' in the environment; a thought representing the property 'P', on the other hand, is a state of a system whose assigned (learned) function is to calibrate the output of the experiential system. Rey (1991) takes both thoughts and experiences to be relations to sentences in the language of thought, and distinguishes them on the basis of (the functional roles of) such sentences' constituent predicates. Lycan (1987, 1996) distinguishes them in terms of their functional-computational profiles. Tye (2000) distinguishes them in terms of their functional roles and the intrinsic structure of their vehicles: thoughts are representations in a language-like medium, whereas experiences are image-like representations consisting of 'symbol-filled arrays.' (The account of mental images in Tye 1991.)
 Phenomenalists tend to make use of the same sorts of features (function, intrinsic structure) in explaining some of the intuitive differences between thoughts and experiences; but they do not suppose that such features exhaust the differences between phenomenal and non-phenomenal representations. For the phenomenalist, it is the phenomenal properties of experiences - qualia themselves - that constitute the fundamental difference between experience and thought. Peacocke (1992), for example, develops the notion of a perceptual 'scenario' (an assignment of phenomenal properties to coordinates of a three-dimensional egocentric space), whose content is 'correct' (a semantic property) if in the corresponding 'scene' (the portion of the external world represented by the scenario) properties are distributed as their phenomenal analogues are in the scenario.
 Another sort of representation championed by phenomenalists (e.g., Block, Chalmers (2003) and Loar (1996)) is the 'phenomenal concept' -, a conceptual/phenomenal hybrid consisting of a phenomenological 'sample' (an image or an occurrent sensation) integrated with (or functioning as) a conceptual component. Phenomenal concepts are postulated to account for the apparent fact (among others) that, as McGinn (1991) puts it, 'you cannot form [introspective] concepts of conscious properties unless you yourself instantiate those properties.' One cannot have a phenomenal concept of a phenomenal property 'P', and, hence, phenomenal beliefs about P, without having experience of 'P', because 'P' itself is (in some way) constitutive of the concept of 'P'. (Jackson 1982, 1986 and Nagel 1974.)
 Though imagery has played an important role in the history of philosophy of mind, the important contemporary literature on it is primarily psychological. In a series of psychological experiments done in the 1970s (summarized in Kosslyn 1980 and Shepard and Cooper 1982), subjects' response time in tasks involving mental manipulation and examination of presented figures was found to vary in proportion to the spatial properties (size, orientation, etc.) of the figures presented. The question of how these experimental results are to be explained has kindled a lively debate on the nature of imagery and imagination.
 Kosslyn (1980) claims that the results suggest that the tasks were accomplished via the examination and manipulation of mental representations that they have spatial properties, i.e., pictorial representations, or images. Others, principally Pylyshyn (1979, 1981, 2003), argue that the empirical facts can be explained in terms exclusively of discursive, or propositional representations and cognitive processes defined over them. (Pylyshyn takes such representations to be sentences in a language of thought.)
 The idea that pictorial representations are literally pictures in the head is not taken seriously by proponents of the pictorial view of imagery.  The claim is, rather, that mental images represent in a way that is relevantly like the way pictures represent. (Attention has been focussed on visual imagery - hence the designation 'pictorial'; Though of course, there may imagery in other modalities - auditory, olfactory, etc. - as well.)
 The distinction between pictorial and discursive representation can be characterized in terms of the distinction between analog and digital representation (Goodman 1976). This distinction has itself been variously understood (Fodor & Pylyshyn 1981, Goodman 1976, Haugeland 1981, Lewis 1971, McGinn 1989), though a widely accepted construal is that analog representation is continuous (i.e., in virtue of continuously variable properties of the representation), while digital representation is discrete (i.e., in virtue of properties a representation either has or doesn't have) (Dretske 1981). (An analog/digital distinction may also be made with respect to cognitive processes. (Block 1983.)) On this understanding of the analog/digital distinction, imaginistic representations, which represent in virtue of properties that may vary continuously (such for being more or less bright, loud, vivid, etc.), would be analog, while conceptual representations, whose properties do not vary continuously (a thought cannot be more or less about Elvis: either it is or it is not) would be digital.
 It might be supposed that the pictorial/discursive distinction is best made in terms of the phenomenal/nonphenomenal distinction, but it is not obvious that this is the case. For one thing, there may be nonphenomenal properties of representations that vary continuously. Moreover, there are ways of understanding pictorial representation that presuppose neither phenomenality nor analogicity. According to Kosslyn (1980, 1982, 1983), a mental representation is 'quasi-pictorial' when every part of the representation corresponds to a part of the object represented, and relative distances between parts of the object represented are preserved among the parts of the representation. But distances between parts of a representation can be defined functionally rather than spatially - for example, in terms of the number of discrete computational steps required to combine stored information about them. (Rey 1981.)
 Tye (1991) proposes a view of images on which they are hybrid representations, consisting both of pictorial and discursive elements. On Tye's account, images are '(labelled) interpreted symbol-filled arrays.' The symbols represent discursively, while their arrangement in arrays has representational significance (the location of each 'cell' in the array represents a specific viewer-centred 2-D location on the surface of the imagined object)
 The contents of mental representations are typically taken to be abstract objects (properties, relations, propositions, sets, etc.). A pressing question, especially for the naturalist, is how mental representations come to have their contents. Here the issue is not how to naturalize content (abstract objects can't be naturalized), but, rather, how to provide a naturalistic account of the content-determining relations between mental representations and the abstract objects they express. There are two basic types of contemporary naturalistic theories of content-determination, causal-informational and functional.
 Causal-informational theories hold that the content of a mental representation is grounded in the information it carries about what does (Devitt 1996) or would (Fodor 1987, 1990) cause it to occur. There is, however, widespread agreement that causal-informational relations are not sufficient to determine the content of mental representations. Such relations are common, but representation is not. Tree trunks, smoke, thermostats and ringing telephones carry information about what they are causally related to, but they do not represent (in the relevant sense) what they carry information about. Further, a representation can be caused by something it does not represent, and can represent something that has not caused it.
 The main attempts to specify what makes a causal-informational state a mental representation are Asymmetric Dependency Theories, the Asymmetric Dependency Theory distinguishes merely informational relations from representational relations on the basis of their higher-order relations to each other: informational relations depend upon representational relations, but not vice-versa. For example, if tokens of a mental state type are reliably caused by horses, cows-on-dark-nights, zebras-in-the-mist and Great Danes, then they carry information about horses, etc. If, however, such tokens are caused by cows-on-dark-nights, etc. because they were caused by horses, but not vice versa, then they represent horses.
 According to Teleological Theories, representational relations are those a representation-producing mechanism has the selected (by evolution or learning) function of establishing. For example, zebra-caused horse-representations do not mean zebra, because the mechanism by which such tokens are produced has the selected function of indicating horses, not zebras. The horse-representation-producing mechanism that responds to zebras is malfunctioning.
 Functional theories, hold that the content of a mental representation are well  grounded in causal computational inferential relations to other mental portrayals other than mental representations.  They differ on whether relata should include all other mental representations or only some of them, and on whether to include external states of affairs. The view that the content of a mental representation is determined by its inferential/computational relations with all other representations is holism; the view it is determined by relations to only some other mental states is localisms (or molecularism). (The view that the content of a mental state depends on none of its relations to other mental states is atomism.) Functional theories that recognize no content-determining external relata have been called solipsistic (Harman 1987). Some theorists posit distinct roles for internal and external connections, the former determining semantic properties analogous to sense, the latter determining semantic properties analogous to reference (McGinn 1982, Sterelny 1989)
 (Reductive) representationalists (Dretske, Lycan, Tye) usually take one or another of these theories to provide an explanation of the (Non-conceptual) content of experiential states. They thus tend to be Externalists (see the next section) about phenomenological as well as conceptual content. Phenomenalists and non-deductive representationalists (Block, Chalmers, Loar, Peacocke, Siewert), on the other hand, take it that the representational content of such states is (at least in part) determined by their intrinsic phenomenal properties. Further, those who advocate a phenomenology-based approach to conceptual content (Horgan and Tiensen, Loar, Pitt, Searle, Siewert) also seem to be committed to internalist individuation of the content (if not the reference) of such states.
 Generally, those who, like informational theorists, think relations to one's (natural or social) environment are (at least partially) determinative of the content of mental representations are Externalists (e.g., Burge 1979, 1986, McGinn 1977, Putnam 1975), whereas those who, like some proponents of functional theories, think representational content is determined by an individual's intrinsic properties alone, are internalists (or individualists).
 This issue is widely taken to be of central importance, since psychological explanation, whether commonsense or scientific, is supposed to be both causal and content-based. (Beliefs and desires cause the behaviours they do because they have the contents they do. For example, the desire that one have a beer and the beliefs that there is beer in the refrigerator and that the refrigerator is in the kitchen may explain one's getting up and going to the kitchen.) If, however, a mental representation's having a particular content is due to factors extrinsic to it, it is unclear how its having that content could determine its causal powers, which, arguably, must be intrinsic. Some who accept the standard arguments for externalism have argued that internal factors determine a component of the content of a mental representation. They say that mental representations have both 'narrow' content (determined by intrinsic factors) and 'wide' or 'broad' content (determined by narrow content plus extrinsic factors). (This distinction may be applied to the sub-personal representations of cognitive science as well as to those of commonsense psychology.
 Narrow content has been variously construed. Putnam (1975), Fodor (1982)), and Block (1986), for example, seem to understand it as something like dedictorial content (i.e., Frégean sense, or perhaps character, à la Kaplan 1989). On this construal, narrow content is context-independent and directly expressible. Fodor (1987) and Block (1986), however, have also characterized narrow content as radically inexpressible. On this construal, narrow content is a kind of proto-content, or content-determinant, and can be specified only indirectly, via specifications of context/wide-content pairings. Both, construe of as a narrow content and are characterized as functions from context to (wide) content. The narrow content of a representation is determined by properties intrinsic to it or its possessor such as its syntactic structure or its intra-mental computational or inferential role (or its phenomenology.
 Burge (1986) has argued that causation-based worries about externalist individuation of psychological content, and the introduction of the narrow notion, are misguided. Fodor (1994, 1998) has more recently urged that a scientific psychology might not need narrow content in order to supply naturalistic (causal) explanations of human cognition and action, since the sorts of cases they were introduced to handle, viz., Twin-Earth cases and Frége cases, are nomologically either impossible or dismissible as exceptions to non-strict psychological laws.
 The leading contemporary version of the Representational Theory of Mind, the Computational Theory of Mind, claims that the brain is a kind of computer and that mental processes are computations. According to the computational theory of mind, cognitive states are constituted by computational relations to mental representations of various kinds, and cognitive processes are sequences of such states. The computational theory of mind and the representational theory of mind, may by attempting to explain all psychological states and processes in terms of mental representation. In the course of constructing detailed empirical theories of human and animal cognition and developing models of cognitive processes' implementable in artificial information processing systems, cognitive scientists have proposed a variety of types of mental representations. While some of these may be suited to be mental relata of commonsense psychological states, some - so-called 'subpersonal' or 'sub-doxastic' representations - are not. Though many philosophers believe that computational theory of mind can provide the best scientific explanations of cognition and behaviour, there is disagreement over whether such explanations will vindicate the commonsense psychological explanations of prescientific representational theory of mind.
 According to Stich's (1983) Syntactic Theory of Mind, for example, computational theories of psychological states should concern themselves only with the formal properties of the objects those states are relations to. Commitment to the explanatory relevance of content, however, is for most cognitive scientists fundamental. That mental processes are computations, which computations are rule-governed sequences of semantically evaluable objects, and that the rules apply to the symbols in virtue of their content, are central tenets of mainstream cognitive science.
 Explanations in cognitive science appeal to a many different kinds of mental representation, including, for example, the 'mental models' of Johnson-Laird 1983, the 'retinal arrays,' 'primal sketches' and '2½ -D sketches' of Marr 1982, the 'frames' of Minsky 1974, the 'sub-symbolic' structures of Smolensky 1989, the 'quasi-pictures' of Kosslyn 1980, and the 'interpreted symbol-filled arrays' of Tye 1991 - in addition to representations that may be appropriate to the explanation of commonsense
Psychological states. Computational explanations have been offered of, among other mental phenomena, belief.
 The classicists hold that mental representations are symbolic structures, which typically have semantically evaluable constituents, and that mental processes are rule-governed manipulations of them that are sensitive to their constituent structure. The connectionists, hold that mental representations are realized by patterns of activation in a network of simple processors ('nodes') and that mental processes consist of the spreading activation of such patterns. The nodes themselves are, typically, not taken to be semantically evaluable; nor do the patterns have semantically evaluable constituents. (Though there are versions of Connectionism -, 'localist' versions - on which individual nodes are taken to have semantic properties (e.g., Ballard 1986, Ballard & Hayes 1984).) It is arguable, however, that localist theories are neither definitive nor representative of the Conceptionist program.
 Classicists are motivated (in part) by properties thought seems to share with language. Jerry Alan Fodor's (1935-), Language of Thought Hypothesis, (Fodor 1975, 1987), according to which the system of mental symbols constituting the neural basis of thought is structured like a language, provides a well-worked-out version of the classical approach as applied to commonsense psychology. According to the language of a thought hypothesis, the potential infinity of complex representational mental states is generated from a finite stock of primitive representational states, in accordance with recursive formation rules. This combinatorial structure accounts for the properties of productivity and systematicity of the system of mental representations. As in the case of symbolic languages, including natural languages (though Fodor does not suppose either that the language of thought hypotheses explains only linguistic capacities or that only verbal creatures have this sort of cognitive architecture), these properties of thought are explained by appeal to the content of the representational units and their combinability into contentful complexes. That is, the semantics of both language and thought is compositional: the content of a complex representation is determined by the contents of its constituents and their structural configuration.
 Connectionists are motivated mainly by a consideration of the architecture of the brain, which apparently consists of layered networks of interconnected neurons. They argue that this sort of architecture is unsuited to carrying out classical serial computations. For one thing, processing in the brain is typically massively parallel. In addition, the elements whose manipulation drive's computation in Conceptionist networks (principally, the connections between nodes) are neither semantically compositional nor semantically evaluable, as they are on the classical approach. This contrast with classical computationalism is often characterized by saying that representation is, with respect to computation, distributed as opposed to local: representation is local if it is computationally basic; and distributed if it is not. (Another way of putting this is to say that for classicists mental representations are computationally atomic, whereas for connectionists they are not.)
 Moreover, connectionists argue that information processing as it occurs in Conceptionist networks more closely resembles some features of actual human cognitive functioning. For example, whereas on the classical view learning involves something like hypothesis formation and testing (Fodor 1981), on the Conceptionist model it is a matter of evolving distribution of 'weight' (strength) on the connections between nodes, and typically does not involve the formulation of hypotheses regarding the identity conditions for the objects of knowledge. The Conceptionist network is 'trained up' by repeated exposure to the objects it is to learn to distinguish; and, though networks typically require many more exposures to the objects than do humans, this seems to model at least one feature of this type of human learning quite well.
 Further, degradation in the performance of such networks in response to damage is gradual, not sudden as in the case of a classical information processor, and hence more accurately models the loss of human cognitive function as it typically occurs in response to brain damage. It is also sometimes claimed that Conceptionist systems show the kind of flexibility in response to novel situations typical of human cognition - situations in which classical systems are relatively 'brittle' or 'fragile.'
 Some philosophers have maintained that Connectionism entails that there are no propositional attitudes. Ramsey, Stich and Garon (1990) have argued that if Conceptionist models of cognition are basically correct, then there are no discrete representational states as conceived in ordinary commonsense psychology and classical cognitive science. Others, however (e.g., Smolensky 1989), hold that certain types of higher-level patterns of activity in a neural network may be roughly identified with the representational states of commonsense psychology. Still others argue that language-of-thought style representation is both necessary in general and realizable within Conceptionist architectures, collect the central contemporary papers in the classicist/Conceptionist debate, and provides useful introductory material as well.
 Whereas Stich (1983) accepts that mental processes are computational, but denies that computations are sequences of mental representations, others accept the notion of mental representation, but deny that computational theory of mind provides the correct account of mental states and processes.
 Van Gelder (1995) denies that psychological processes are computational. He argues that cognitive systems are dynamic, and that cognitive states are not relations to mental symbols, but quantifiable states of a complex system consisting of (in the case of human beings) a nervous system, a body and the environment in which they are embedded. Cognitive processes are not rule-governed sequences of discrete symbolic states, but continuous, evolving total states of dynamic systems determined by continuous, simultaneous and mutually determining states of the systems components. Representation in a dynamic system is essentially information-theoretic, though the bearers of information are not symbols, but state variables or parameters.
 Horst (1996), on the other hand, argues that though computational models may be useful in scientific psychology, they are of no help in achieving a philosophical understanding of the intentionality of commonsense mental states. Computational theory of mind attempts to reduce the intentionality of such states to the intentionality of the mental symbols they are relations to. But, Horst claims, the relevant notion of symbolic content is essentially bound up with the notions of convention and intention. So the computational theory of mind involves itself in a vicious circularity: the very properties that are supposed to be reduced are (tacitly) appealed to in the reduction.
 To say that a mental object has semantic properties is, paradigmatically, to say that it may be about, or be true or false of, an object or objects, or that it may be true or false simpiciter. Suppose I think that you took to sniffing snuff. I am thinking about you, and if what I think of you (that they take snuff) is true of you, then my thought is true. According to representational theory of mind such states are to be explained as relations between agents and mental representations. To think that you take snuff is to token in some way a mental representation whose content is that ocelots take snuff. On this view, the semantic properties of mental states are the semantic properties of the representations they are relations to.
 Linguistic acts seem to share such properties with mental states. Suppose I say that you take snuff. I am talking about you, and if what I say of you (that they take snuff) is true of them, then my utterance is true. Now, to say that you take snuff is (in part) to utter a sentence that means that you take snuff. Many philosophers have thought that the semantic properties of linguistic expressions are inherited from the intentional mental states they are conventionally used to express. On this view, the semantic properties of linguistic expressions are the semantic properties of the representations that are the mental relata of the states they are conventionally used to express.
 It is also widely held that in addition to having such properties as reference, truth-conditions and truth - so-called extensional properties - expressions of natural languages also have intensional properties, in virtue of expressing properties or propositions - i.e., in virtue of having meanings or senses, where two expressions may have the same reference, truth-conditions or truth value, yet express different properties or propositions (Frége 1892/1997). If the semantic properties of natural-language expressions are inherited from the thoughts and concepts they express (or vice versa, or both), then an analogous distinction may be appropriate for mental representations.
Philosophers and scientists alike, have often held that the simplicity or parsimony of a theory is one reason, all else being equal, to view it as true. This goes beyond the unproblematic idea that simpler theories are easier to work with and gave greater aesthetic appeal.
 One theory is more parsimonious than another when it postulates fewer entities, processes, changes or explanatory principles: The simplicity of a theory depends on essentially the same consecrations, though parsimony and simplicity obviously become the same. Demanding clarification of what makes one theory simpler or more parsimonious is plausible than another before the justification of these methodological maxims can be addressed.
 If we set this description problem to one side, the major normative problem is as follows: What reason is there to think that simplicity is a sign of truth? Why should we accept a simpler theory instead of its more complex rivals? Newton and Leibniz thought that the answer was to be found in a substantive fact about nature. In Principia, Newton laid down as his first Rule of Reasoning in Philosophy that nature does nothing in vain . . . for Nature is pleased with simplicity and affects not the pomp of superfluous causes. Leibniz hypothesized that the actual world obeys simple laws because Gods' taste for simplicity influenced his decision about which world to actualize.
 The tragedy of the Western mind, described by Koyré, is a direct consequence of the stark Cartesian division between mind and world. We discovered the certain principles of physical reality, said Descartes, not by the prejudices of the senses, but by the light of reason, and which thus possess so great evidence that we cannot doubt of their truth. Since the real, or that which actually exists external to ourselves, was in his view only that which could be represented in the quantitative terms of mathematics, Descartes conclude that all quantitative aspects of reality could be traced to the deceitfulness of the senses.
 The most fundamental aspect of the Western intellectual tradition is the assumption that there is a fundamental division between the material and the immaterial world or between the realm of matter and the realm of pure mind or spirit. The metaphysical frame-work based on this assumption is known as ontological dualism.  As the word dual implies, the framework is predicated on ontology, or a conception of the nature of God or being, that assumes reality has two distinct and separable dimensions. The concept of Being as continuous, immutable and having a prior or separate existence from the world of change dates from the ancient Greek philosopher Parmenides. The same qualities were associated with the God of the Judeo-Christian tradition, and they were considerably amplified by the role played in theology by Platonic and Neo-Platonism philosophy.
 Nicolas Copernicus, Galileo, Johannes Kepler, and Isaac Newton were all inheritors of a cultural tradition in which ontological dualism was a primary article of faith. Hence the idealization of the mathematical ideal as a source of communion with God, which dates from Pythagoras, provided a metaphysical foundation for the emerging natural sciences. This explains why, the creators of classical physics believed that doing physics was a form of communion with the geometrical and mathematical forms resident in the perfect mind of God. This view would survive in a modified form in what is now known as Einsteinian epistemology and accounts in no small part for the reluctance of many physicists to accept the epistemology associated with the Copenhagen Interpretation.
 At the beginning of the nineteenth century, Pierre-Simon LaPlace, along with a number of other French mathematicians, advanced the view that the science of mechanics constituted a complete view of nature. Since this science, by observing its epistemology, had revealed itself to be the fundamental science, the hypothesis of God was, they concluded, entirely unnecessary.
 LaPlace is recognized for eliminating not only the theological component of classical physics but the entire metaphysical component as well. The epistemology of science requires, he said, that we proceed by inductive generalizations from observed facts to hypotheses that are tested by observed conformity of the phenomena. What was unique about LaPlaces view of hypotheses was his insistence that we cannot attribute reality to them. Although concepts like force, mass, motion, cause, and laws are obviously present in classical physics, they exist in LaPlaces view only as quantities. Physics is concerned, he argued, with quantities that we associate as a matter of convenience with concepts, and the truth about nature are only the quantities.
 As this view of hypotheses and the truth of nature as quantities was extended in the nineteenth century to a mathematical description of phenomena like heat, light, electricity, and magnetism. LaPlace’s assumptions about the actual character of scientific truth seemed correct. This progress suggested that if we could remove all thoughts about the nature of or the source of phenomena, the pursuit of strictly quantitative concepts would bring us to a complete description of all aspects of physical reality. Subsequently, figures like Comte, Kirchhoff, Hertz, and Poincaré developed a program for the study of nature hat was quite different from that of the original creators of classical physics.
 The seventeenth-century view of physics as a philosophy of nature or as natural philosophy was displaced by the view of physics as an autonomous science that was the science of nature. This view, which was premised on the doctrine of positivism, promised to subsume all of nature with a mathematical analysis of entities in motion and claimed that the true understanding of nature was revealed only in the mathematical description. Since the doctrine of positivism assumes that the knowledge we call physics resides only in the mathematical formalism of physical theory, it disallows the prospect that the vision of physical reality revealed in physical theory can have any other meaning. In the history of science, the irony is that positivism, which was intended to banish metaphysical concerns from the domain of science, served to perpetuate a seventeenth-century metaphysical assumption about the relationship between physical reality and physical theory.
 Epistemology since Hume and Kant has drawn back from this theological underpinning. Indeed, the very idea that nature is simple (or uniform) has come in for a critique. The view has taken hold that a preference for simple and parsimonious hypotheses is purely methodological: It is constitutive of the attitude we call scientific and makes no substantive assumption about the way the world is.
 A variety of otherwise diverse twentieth-century philosophers of science have attempted, in different ways, to flesh out this position. Two examples must suffice here: Hesse (1969) as, for summaries of other proposals. Popper (1959) holds that scientists should prefer highly falsifiable (improbable) theories: He tries to show that simpler theories are more falsifiable, also Quine (1966), in contrast, sees a virtue in theories that are highly probable, he argues for a general connexion between simplicity and high probability.
 Both these proposals are global. They attempt to explain why simplicity should be part of the scientific method in a way that spans all scientific subject matters. No assumption about the details of any particular scientific problem serves as a premiss in Popper or Quine's arguments.
 Newton and Leibniz thought that the justification of parsimony and simplicity flows from the hand of God: Popper and Quine try to justify these methodologically median of importance is without assuming anything substantive about the way the world is. In spite of these differences in approach, they have something in common. They assume that all users of parsimony and simplicity in the separate sciences can be encompassed in a single justifying argument. That recent developments in confirmation theory suggest that this assumption should be scrutinized. Good (1983) and Rosenkrantz (1977) has emphasized the role of auxiliary assumptions in mediating the connexion between hypotheses and observations. Whether a hypothesis is well supported by some observations, or whether one hypothesis is better supported than another by those observations, crucially depends on empirical background assumptions about the inference problem here. The same view applies to the idea of prior probability (or, prior plausibility). In of a single hypo-physical science if chosen as an alternative to another even though they are equally supported by current observations, this must be due to an empirical background assumption.
 Principles of parsimony and simplicity mediate the epistemic connexion between hypotheses and observations. Perhaps these principles are able to do this because they are surrogates for an empirical background theory. It is not that there is one background theory presupposed by every appeal to parsimony; this has the quantifier order backwards. Rather, the suggestion is that each parsimony argument is justified only to each degree that it reflects an empirical background theory about the subjective matter. On this theory is brought out into the open, but the principle of parsimony is entirely dispensable (Sober, 1988).
 This local approach to the principles of parsimony and simplicity resurrects the idea that they make sense only if the world is one way rather than another. It rejects the idea that these maxims are purely methodological. How defensible this point of view is, will depend on detailed case studies of scientific hypothesis evaluation and on further developments in the theory of scientific inference.
 It is usually not found of one and the same that, an inference is a (perhaps very complex) act of thought by virtue of which act (1) I pass from a set of one or more propositions or statements to a proposition or statement and (2) it appears that the latter are true if the former is or are. This psychological characterization has occurred over a wider summation of literature under lesser than inessential variations. Desiring a better characterization of inference is natural. Yet attempts to do so by constructing a fuller psychological explanation fail to comprehend the grounds on which inference will be objectively valid-A point elaborately made by Gottlob Frége. Attempts to understand the nature of inference through the device of the representation of inference by formal-logical calculations or derivations better (1) leave us puzzled about the relation of formal-logical derivations to the informal inferences they are supposedly to represent or reconstruct, and (2) leaves us worried about the sense of such formal derivations. Are these derivations inference? Are not informal inferences needed in order to apply the rules governing the constructions of formal derivations (inferring that this operation is an application of that formal rule)? These are concerns cultivated by, for example, Wittgenstein.
 Coming up with an adequate characterized inference, and even working out what would count as a very adequate characterization here is demandingly by no means nearly some resolved philosophical problem.
 Traditionally, a proposition that is not a conditional, as with the affirmative and negative, modern opinion is wary of the distinction, since what appears categorical may vary with the choice of a primitive vocabulary and notation. Apparently categorical propositions may also turn out to be disguised conditionals: X is intelligent (categorical?) Equivalent, if X is given a range of tasks, she does them better than many people (conditional?). The problem is not merely one of classification, since deep metaphysical questions arise when facts that seems to be categorical and therefore solid, come to seem by contrast conditional, or purely hypothetical or potential.
 Its condition of some classified necessity is so proven sufficient that if p is a necessary condition of q, then q cannot be true unless p; is true? If p is a sufficient condition, thus steering well is a necessary condition of driving in a satisfactory manner, but it is not sufficient, for one can steer well but drive badly for other reasons. Confusion may result if the distinction is not heeded. For example, the statement that A causes B may be interpreted to mean that A is itself a sufficient condition for B, or that it is only a necessary condition fort B, or perhaps a necessary parts of a total sufficient condition. Lists of conditions to be met for satisfying some administrative or legal requirement frequently attempt to give individually necessary and jointly sufficient sets of conditions.
 What is more that if any proposition of the form if p then q. The condition hypothesized p. Is called the antecedent of the conditionals, and q, the consequent? Various kinds of conditional have been distinguished.  Its weakest is that of material implication, merely telling that either not-p or q. stronger conditionals include elements of modality, corresponding to the thought that if p is truer then q must be true. Ordinary language is very flexible in its use of the conditional form, and there is controversy whether conditionals are better treated semantically,  yielding differently finds of conditionals with different meanings, or pragmatically, in which case there should be one basic meaning with surface differences arising from other implicatures.
 It follows from the definition of strict implication that a necessary proposition is strictly implied by any proposition, and that an impossible proposition strictly implies any proposition. If strict implication corresponds to q follows from p, then this means that a necessary proposition follows from anything at all, and anything at all follows from an impossible proposition. This is a problem if we wish to distinguish between valid and invalid arguments with necessary conclusions or impossible premises.
 The Humean problem of induction is that if we would suppose that there is some property 'A' concerning and observational or an experimental situation, and that out of a large number of observed instances of 'A', some fraction m/n (possibly equal to 1) has also been instances of some logically independent property 'B'. Suppose further that the background proportionate circumstances not specified in these descriptions have been varied to a substantial degree and that there is no collateral information available concerning the frequency of 'B's' among 'A's' or concerning causal or nomologically connections between instances of 'A' and instances of 'B'.
 In this situation, an enumerative or instantial induction inference would move rights from the premise, that m/n of observed 'A's' are 'B's' to the conclusion that approximately m/n of all 'A's' are 'B's'. (The usual probability qualification will be assumed to apply to the inference, rather than being part of the conclusion.) Here the class of 'A's' should be taken to include not only unobserved 'A's' and future 'A's', but also possible or hypothetical 'A's' (an alternative conclusion would concern the probability or likelihood of the adjacently observed 'A' being a 'B').
 The traditional or Humean problem of induction, often referred to simply as the problem of induction, is the problem of whether and why inferences that fit this schema should be considered rationally acceptable or justified from an epistemic or cognitive standpoint, i.e., whether and why reasoning in this way is likely to lead to true claims about the world. Is there any sort of argument or rationale that can be offered for thinking that conclusions reached in this way are likely to be true in the corresponding premises is true - or even that their chances of truth are significantly enhanced?
 Humes discussion of this issue deals explicitly only with cases where all observed 'A's' are 'B's' and his argument applies just as well to the more general case. His conclusion is entirely negative and sceptical: Inductive inferences are not rationally justified, but are instead the result of an essentially a-rational process, custom or habit. Hume (1711-76) challenges the proponent of induction to supply a cogent lignin of reasoning that leads from an inductive premise to the corresponding conclusion and offers an extremely influential argument in the form of a dilemma (a few times referred to as Humes fork), that either our actions are determined, in which case we are not responsible for them, or they are the result of random events, under which case we are also not responsible for them.
 Such reasoning would, he argues, have to be either deductively demonstrative reasoning in the concerning relations of ideas or experimental, i.e., empirical, that reasoning concerning matters of fact or existence. It cannot be the former, because all demonstrative reasoning relies on the avoidance of contradiction, and it is not a contradiction to suppose that the course of nature may change, that an order that was observed in the past and not of its continuing against the future: But it cannot be, as the latter, since any empirical argument would appeal to the success of such reasoning about an experience, and the justifiability of generalizing from experience are precisely what is at issue-so that any such appeal would be question-begging. Hence, Hume concludes that there can be no such reasoning (1748).
 An alternative version of the problem may be obtained by formulating it with reference to the so-called Principle of Induction, which says roughly that the future will resemble the past or, somewhat better, that unobserved cases will resemble observed cases. An inductive argument may be viewed as enthymematic, with this principle serving as a supposed premiss, in which case the issue is obviously how such a premiss can be justified. Hume's argument is then that no such justification is possible: The principle cannot be justified a prior because having possession of been true in experiences without obviously begging the question is not contradictory to have possession of been true in experiences without obviously begging the question.
 The predominant recent responses to the problem of induction, at least in the analytic tradition, in effect accept the main conclusion of Humes argument, namely, that inductive inferences cannot be justified in the sense of showing that the conclusion of such an inference is likely to be true if the premise is true, and thus attempt to find another sort of justification for induction. Such responses fall into two main categories: (I) Pragmatic justifications or vindications of induction, mainly developed by Hans Reichenbach (1891-1953), and (ii) ordinary language justifications of induction, whose most important proponent is Frederick, Peter Strawson (1919- ). In contrast, some philosophers still attempt to reject Humes dilemma by arguing either (iii) That, contrary to appearances, induction can be inductively justified without vicious circularity, or (iv) that an anticipatory justification of induction is possible after all. In that:
   (1) Reichenbachs view is that induction is best regarded, not as a form of inference, but rather as a method for arriving at posits regarding, i.e., the proportion of what remain besides such a speculative assertion would not claim asserted to be true, but is instead an intellectual wager analogous to a bet made by a gambler. Understood in this way, the inductive method says that one should posit that the observed proportion is, within some measure of an approximation, the true proportion and then continually correct that initial posit as new information comes in.
 The gamblers bet is normally an appraised posit, i.e., he knows the chances or odds that the outcome on which he bets will actually occur. In contrast, the inductive bet is a blind posit: We do not know the chances that it will succeed or even that success is that it will succeed or even that success is possible. What we are gambling on when we make such a bet is the value of a certain proportion in the independent world, which Reichenbach construes as the limit of the observed proportion as the number of cases increases to infinity. Nevertheless, we have no way of knowing that there are even such a limit, and no way of knowing that the proportion of As are in addition of convergence, the end on some stable value than varying at random. If we cannot know that this limit exists, then we obviously cannot know that we have any definite chance of finding it.
 What we can know, according to Reichenbach, is that if there is a truth of this sort to be found, the inductive method will eventually find it. That this is so is an analytic consequence of Reichenbachs account of what it is for such a limit to exist. The only way that the inductive method of making an initial posit and then refining it in light of new observations can fail eventually to arrive at the true proportion is if the series of observed proportions never converges on any stable value, which means that there is no truth to be found pertaining the proportion of an additionally make up of an induction if justified, not by showing that it will succeed or indeed, that it has any definite likelihood of success, but only by showing that it will succeed if success is possible. Reichenbachs claim is that no more than this can be established for any method, and hence that induction gives us our best chance for success, our best gamble in a situation where there is no alternative to gambling.
 This pragmatic response to the problem of induction faces several serious problems. First, there are indefinitely many other methods for arriving at posits for which the same sort of defence can be given-methods that yield the same result as the inductive method over time but differ arbitrarily before long. Despite the efforts of others, it is unclear that there is any satisfactory way to exclude such alternatives, in order to avoid the result that any arbitrarily chosen short-term posit is just as reasonable as the inductive posit. Second, even if there is a truth of the requisite sort to be found, the inductive method is only guaranteed to find it or even to come within any specifiable distance of it in the indefinite long run. All the same, any actual application of inductive results always takes place in the presence to the future eventful states in making the relevance of the pragmatic justification to actual practice uncertainly. Third, and most important, it needs to be emphasized that Reichenbachs response to the problem simply accepts the claim of the Humean sceptic that an inductive premise never provides the slightest reason for thinking that the corresponding inductive conclusion is true. Reichenbach himself is quite candid on this point, but this does not alleviate the intuitive implausibility of saying that we have no more reason for thinking that our scientific and commonsense conclusions that result in the induction of it  . . .  is true than, to use Reichenbachs own analogy (1949), a blind man wandering in the mountains who feels an apparent trail with his stick has for thinking that following it will lead him to safety.
 An approach to induction resembling Reichenbachs claiming in that those particular inductive conclusions are posits or conjectures, than the conclusions of cogent inferences, is offered by Popper. However, Poppers view is even more overtly sceptical: It amounts to saying that all that can ever be said in favours of the truth of an inductive claim is that the claim has been tested and not yet been shown to be false.
   (2) The ordinary language response to the problem of induction has been advocated by many philosophers, none the less, Strawson claims that the question whether induction is justified or reasonable makes sense only if it tacitly involves the demand that inductive reasoning meet the standards appropriate to deductive reasoning, i.e., that the inductive conclusions are shown to follow deductively from the inductive assumption. Such a demand cannot, of course, be met, but only because it is illegitimate: Inductive and deductive reasons are simply fundamentally different kinds of reasoning, each possessing its own autonomous standards, and there is no reason to demand or expect that one of these kinds meet the standards of the other. Whereas, if induction is assessed by inductive standards, the only ones that are appropriate, then it is obviously justified.
 The problem here is to understand to what this allegedly obvious justification of an induction amount. In his main discussion of the point (1952), Strawson claims that it is an analytic true statement that believing it a conclusion for which there is strong evidence is reasonable and an analytic truth that inductive evidence of the sort captured by the schema presented earlier constitutes strong evidence for the corresponding inductive conclusion, thus, apparently yielding the analytic conclusion that believing it a conclusion for which there is inductive evidence is reasonable. Nevertheless, he also admits, indeed insists, that the claim that inductive conclusions will be true in the future is contingent, empirical, and may turn out to be false (1952). Thus, the notion of reasonable belief and the correlative notion of strong evidence must apparently be understood in ways that have nothing to do with likelihood of truth, presumably by appeal to the standard of reasonableness and strength of evidence that are accepted by the community and are embodied in ordinary usage.
 Understood in this way, Strawsons response to the problem of inductive reasoning does not speak to the central issue raised by Humean scepticism: The issue of whether the conclusions of inductive arguments are likely to be true. It amounts to saying merely that if we reason in this way; we can correctly call ourselves reasonable and our evidence strong, according to our accepted community standards. Nevertheless, to the undersealing of issue of whether following these standards is a good way to find the truth; the ordinary language response appears to have nothing to say.
   (3) The main attempts to show that induction can be justified inductively have concentrated on showing that such as a defence can avoid circularity. Skyrms (1975) formulate, perhaps the clearest version of this general strategy. The basic idea is to distinguish different levels of inductive argument: A first level in which induction is applied to things other than arguments: A second level in which it is applied to arguments at the first level, arguing that they have been observed to succeed so far and hence are likely to succeed in general: A third level in which it is applied in the same way to arguments at the second level, and so on. Circularity is allegedly avoided by treating each of these levels as autonomous and justifying the argument at each level by appeal to an argument at the next level.
 One problem with this sort of move is that even if circularity is avoided, the movement to Higher and Higher levels will clearly eventually fail simply for lack of evidence: A level will reach at which there have been enough successful inductive arguments to provide a basis for inductive justification at the next Higher level, and if this is so, then the whole series of justifications collapses. A more fundamental difficulty is that the epistemological significance of the distinction between levels is obscure. If the issue is whether reasoning in accord with the original schema offered above ever provides a good reason for thinking that the conclusion is likely to be true, then it still seems question-begging, even if not flatly circular, to answer this question by appeal to anther argument of the same form.
   (4) The idea that induction can be justified on a pure priori basis is in one way the most natural response of all: It alone treats an inductive argument as an independently cogent piece of reasoning whose conclusion can be seen rationally to follow, although perhaps only with probability from its premise. Such an approach has, however, only rarely been advocated (Russell, 19132 and BonJour, 1986), and is widely thought to be clearly and demonstrably hopeless.
 Many on the reasons for this pessimistic view depend on general epistemological theses about the possible or nature of anticipatory cognition. Thus if, as Quine alleges, there is no a prior justification of any kind, then obviously a prior justification for induction is ruled out. Or if, as more moderate empiricists have in claiming some preexistent knowledge should be analytic, then again a prevenient justification for induction seems to be precluded, since the claim that if an inductive premise is truer, then the conclusion is likely to be true does not fit the standard conceptions of analyticity. A consideration of these matters is beyond the scope of the present spoken exchange.
 There are, however, two more specific and quite influential reasons for thinking that an early approach is impossible that can be briefly considered, first, there is the assumption, originating in Hume, but since adopted by very many of others, that a move forward in  the defence of induction would have to involve turning induction into deduction, i.e., showing, per impossible, that the inductive conclusion follows deductively from the premise, so that it is a formal contradiction to accept the latter and deny the former. However, it is unclear why a prior approach need be committed to anything this strong. It would be enough if it could be argued that it is deductively unlikely that such a premise is true and corresponding conclusion false.
 Second, Reichenbach defends his view that pragmatic justification is the best that is possible by pointing out that a completely chaotic world in which there is simply not true conclusion to be found as to the proportion of, in addition that occur of, but is neither impossible nor unlikely from a purely a prior standpoint, the suggestion being that therefore there can be no a prior reason for thinking that such a conclusion is true. Nevertheless, there is still a substring way in laying that a chaotic world is a prior neither impossible nor unlikely without any further evidence does not show that such a world is not a prior unlikely and a world containing such-and-such regularity might anticipatorily be somewhat likely in relation to an occurrence of a long running pattern of evidence in which a certain stable proportion of observed events -. An occurrence, it might be claimed, that would be highly unlikely in a chaotic world (BonJour, 1986).
 Goodman's new riddle of induction purports that we suppose that before some specific time t (perhaps the year 2000) we observe a larger number of emeralds (property A) and find them all to be green (property B). We proceed to reason inductively and conclude that all emeralds are green Goodman points out, however, that we could have drawn a quite different conclusion from the same evidence. If we define the term stuff to mean green if examined before t and blue examined after t ?, then all of our observed emeralds will also be gruing.  A parallel inductive argument will yield the conclusion that all emeralds are gruing, and hence that all those examined after the year 2000 will be blue. Presumably the first of these concisions is genuinely supported by our observations and the second is not.  Nevertheless, the problem is to say why this is so and to impose some further restriction upon inductive reasoning that will permit the first argument and exclude the second.
 The obvious alternative suggestion is that stuff.  Similar predicates do not correspond to genuine, purely qualitative properties in the way that green and blueness does, and that this is why inductive arguments involving them are unacceptable. Goodman, however, claims to be unable to make clear sense of this suggestion, pointing out that the relations of formal desirability are perfectly symmetrical: Stuff may be defined in terms if, green and blue, but green an equally well be defined in terms of stuff and green (blue if examined before t and green if examined after t).
 The stuff that has been recognized from its complicated and most puzzling of named paradoxes that only demonstrate the importance of categorization, in that sometimes it is itemized as gruing, if examined of a presence to the future, before future time t and green, or not so examined and blue. Even though all emeralds in our evidence class stuff, we ought to must infer that all emeralds are gruing. For stuff is unpredictable, and cannot transmit credibility from known to unknown cases. Only projectable predicates are right for induction. Goodman considers entrenchment the key to projectibility having a long history of successful protection, stuff is entrenched, lacking such a history, stuff is not. A hypothesis is projectable, Goodman suggests, only if its predicates (or suitable related ones) are much better entrenched than its rivalrous past successes that do not assume future ones. Induction remains a risky business. The rationale for favouring entrenched predicates is pragmatic. Of the possible projections from our evidence class, the one that fits with past practices enables us to utilize our cognitive resources best. Its prospects of being true are worse than its competitors and its cognitive utility is greater.
 So, to a better understanding of induction we should then linearize its term for which is most widely used for any process of reasoning that takes us from empirical premises to empirical conclusions supported by the premises, but not deductively entailed by them. Inductive arguments are therefore kinds of applicative arguments, in which something beyond the content of the premise is inferred as probable or supported by them. Induction is, however, commonly distinguished from arguments to theoretical explanations, which share this applicative character, by being confined to inferences in which he conclusion involves the same properties or relations as the premises. The central example is induction by simple enumeration, where from premises telling that Fa, Fb, Fc . . . where 'a', 'b', 'C's', are all of some kind 'G', it is inferred that 'G's' from outside the sample, such as future 'G's', will be 'F', or perhaps that all 'G's' are 'F'. In this, which and the other persons deceive them, children may infer that everyone is a deceiver: Different, but similar inferences of a property by some object to the same objects future possession of the same property, or from the constancy of some law-like pattern in events and states of affairs ti its future constancy. All objects we know of attract each other with a force inversely proportional to the square of the distance between them, so perhaps they all do so, and will always do so.
 The rational basis of any inference was challenged by Hume, who believed that induction presupposed belief in the uniformity of nature, but that this belief has no defence in reason, and merely reflected a habit or custom of the mind. Hume was not therefore sceptical about the role of reason in either explaining it or justifying it. Trying to answer Hume and to show that there is something rationally compelling about the inference referred to as the problem of induction. It is widely recognized that any rational defence of induction will have to partition well-behaved properties for which the inference is plausible (often called projectable properties) from badly behaved ones, for which it is not. It is also recognized that actual inductive habits are more complex than those of similar enumeration, and that both common sense and science pay attention to such giving factors as variations within the sample giving us the evidence, the application of ancillary beliefs about the order of nature, and so on.
 Nevertheless, the fundamental problem remains that and experience condition by application show us only events occurring within a very restricted part of a vast spatial and temporal order about which we then come to believe things.
 Uncompounded by its belonging of a confirmation theory finding of the measure to which evidence supports a theory fully formalized confirmation theory would dictate the degree of confidence that a rational investigator might have in a theory, given some-body of evidence. The grandfather of confirmation theory is Gottfried Leibniz (1646-1718), who believed that a logically transparent language of science would be able to resolve all disputes. In the 20th century a fully formal confirmation theory was a main goal of the logical positivist, since without it the central concept of verification by empirical evidence itself remains distressingly unscientific. The principal developments were due to Rudolf Carnap (1891-1970), culminating in his Logical Foundations of Probability (1950). Carnaps idea was that the measure necessitated would be the proportion of logically possible states of affairs in which the theory and the evidence both hold, compared ti the number in which the evidence itself holds that the probability of a preposition, relative to some evidence, is a proportion of the range of possibilities under which the proposition is true, compared to the total range of possibilities left by the evidence. The difficulty with the theory lies in identifying sets of possibilities so that they admit of measurement. It therefore demands that we can put a measure on the range of possibilities consistent with theory and evidence, compared with the range consistent with the evidence alone.
 Among the obstacles the enterprise meets, is the fact that while evidence covers only a finite range of data, the hypotheses of science may cover an infinite range. In addition, confirmation proves to vary with the language in which the science is couched, and the Carnapian programme has difficulty in separating genuinely confirming variety of evidence from less compelling repetition of the same experiment. Confirmation also proved to be susceptible to acute paradoxes. Finally, scientific judgment seems to depend on such intangible factors as the problems facing rival theories, and most workers have come to stress instead the historically situated scene of what would appear as a plausible distinction of a scientific knowledge at a given time.