It is plausible to assume that all that is required for semantic knowledge, conceived as knowledge which truth condition has been stated, is that the following two conditions be met. First, a sufficient number of current or former (late) members of the speech community to which the speaker belongs possess appropriate background know-how. Second, the speaker stands in an appropriate social relation to these members (a relation that would enable the speaker to express communal knowledge by a true sentence whose content she may be unable to “verify” herself).
These members are experts; they are capable of “verifying” or “falsifying” the semantic content of the sentence in question, as opposed to merely grasping it. Other members of their social group participate in their knowledge thanks to intersubjective processes of information transfer. The main idea behind this approach is an adaptation of Evans’ distinction between what he calls name-producers and mere name-consumers, which is used to substantiate the above distinction between two kinds of knowledge regarding a sentence's truth condition.[16] This strategy leads to a social-epistemological conception of the background of meaning and to a version of contextualism that preserves basic insights of anti-contextualists like Borg. Evans writes:
Let us consider an ordinary proper-name-using practice, in which the name ‘NN’ is used to refer to the person x. The distinctive mark of any such practice is the existence of a core group of speakers who have been introduced to the practice via their acquaintance with x. They have on some occasion been told, or anyway have come to learn, a truth which they could then express as ‘This is NN’, where ‘This’ makes a demonstrative reference to x. Once a speaker has learned such a truth, the capacity to re-identify persons over time enables him to recognize later occasions on which the judgement ‘This is NN’ may be made, and hence in connection with which the name ‘NN’ may be used. […] Members of this core group, whom I shall call ‘producers’ […], do more than merely use the name to refer to x; they have dealings with x from time to time, and use the name in those dealings – they know x, and further, they know x as NN. […] [T]he expression does not become a name for x unless it has a certain currency among those who know x – only then can we say that x is known as NN. […] Perhaps in the early stages of its existence all the participants in the name-using practice will be producers, but this is unlikely to remain so for long. Others, who are not acquainted with x, can be introduced into the practice, either by helpful explanations of the form ‘NN is the φ’, or just by hearing sentences in which the name is used. I shall call these members ‘consumers’, since on the whole they are not able to inject new information into the practice, but must rely upon the information-gathering transactions of the producers. […] Let us now consider the last phase of a practice of a name-using practice, when all the participants are consumers. […]. Later consumers manifest the intention to be participating in this practice, and, using a name which, in the practice, refers to Livingstone, themselves refer to Livingstone. Thus the practice is maintained with a constant reference, perhaps for very long periods of time. (Evans 1982, pp. 376-393)
If we adapt Evans’ distinction between two types of name-users for present purposes, we can say that in a given community there have to be, or have to have been (see the last three sentences of the quotation), people “in the know” regarding (what we use to call) the red colour of apples, or regarding a particular practice of opening mountains, grass, or the sun, in order for the sentences S2–S4 to be candidates, in virtue of their literal meaning, for the expression of knowledge available to us through these sentences.[17] There have to be (current or late) “producers” in order for these sentences to express a semantic content determining truth conditions, thus displaying a clear, interpretable respective meaning in that linguistic community—and this requires that the sentences have a community-wide usage upheld by recourse to (current or late) producers. They know (or knew) how to “verify” the respective meaning of assertive utterances of the sentence, in the relevant usage; i.e., they know which fact (if any) would make it the case that the truth condition is satisfied; they know how to follow a corresponding order, and so on.
The rest of the speech community merely knows the truth condition and can gain and transfer information an utterance of the sentence bears without themselves being in the know–that is, without having the original knowledge only the producers have in their possession. They may acquire and transfer knowledge (sometimes) by testimony, thanks to the existence of a community-wide practice of sentence-usage sustained by intersubjective processes of information transfer, in a way yet to be understood in more detail.
Eventually, mere consumers “must rely upon the information-gathering transactions of the producers,” to use Evans' formulation. Mere consumers have semantic knowledge, but they lack more substantive knowledge. Semantics is concerned with the content of their semantic knowledge. Mere consumers need a background of what Searle calls social practices, including social practices of language use. However, they lack the producers’ individual or personal background know-how and thus their substantive knowledge regarding truth conditions, which requires such know-how (i.e., the knowledge of how to “verify” those conditions).
What kind of individual background do the producers need in order to be able to make possible social practices of language use that allow all members of their speech community to express and grasp semantic contents determining particular truth conditions? In his 1978 paper, which some regard as the constitutive document of contextualism, Searle stresses the importance of background assumptions, such as the assumption that there is a field of gravitation or that things offer resistance to pressure, which is usually taken for granted, quite unreflectedly, when we speak about middle-sized everyday objects such as apples and boxes. This may at first sound like the requirement of what might be called background knowledge, consisting of intentional states, i.e., certain epistemically distinguished beliefs. However, especially in his later writings, Searle stresses the non-intentional character of the background, characterizing it as consisting of non-intentional capacities—which I have referred to above as background know-how, and which would include the ability to perform social practices. Searle has formulated a thesis about the relation between intentionality, on the one hand, and background know-how on the other, a thesis he calls the “hypothesis of the Background”:
Another way to state [the hypothesis of the Background] is to say that all representation, whether in language, thought, or experience, only succeeds given a set of nonrepresentational capacities. In my technical jargon, intentional states only determine conditions of satisfaction relative to a set of capacities that are not themselves intentional. (Searle 1992, p. 175)
Later in the same book chapter he explains:
The actual content [sc. of an intentional state] is insufficient to determine the conditions of satisfaction. […] Even if you spell out all contents of the mind as a set of conscious rules, thoughts, beliefs, etc., you still require a set of Background capacities for their interpretation. (Searle 1992, pp. 189-190)
This addition to the formulation in the penultimate quotation is, I think, false—or even absurd. The respective meaning of an utterance is the intentional content of the mental state given voice to in the narrow sense, which means that intentional content is precisely what determines the truth condition (or, more generally, the conditions of satisfaction). Indeed, Searle seems to agree:
[…] I want to capture our ordinary intuition that the man who has the belief that Sally cut the cake has a belief with exactly the same propositional content as the literal assertion ‘Sally cut the cake.’ (Searle 1992, p. 184)
I take it to be a definitional truth that intentional content provides the answer to Wittgenstein's question “What makes my representation of him a representation of him?”. A conception of intentional content must spell out this answer. It makes no sense to conceive intentional content along the lines of Searle’s addition in the penultimate quotation, just as it makes no sense (pace Searle) to say of semantic content, properly construed, that it is not self-applying, or that it needs to be interpreted against a non-representational background in order to determine reference or satisfaction conditions.
In the following passage Searle commits himself to radical contextualism:
An utterance of [the sentence ‘Sally gave John the key, and he opened the door’] would normally convey that first Sally gave John the key, and later he opened the door, and that he opened the door with the key. There is much discussion about the mechanisms by which this additional content is conveyed, given that it is not encoded in the literal meaning of the sentence. The suggestion, surely correct, is that sentence meaning, at least to a certain extent, underdetermines what the speaker says when he utters the sentence. Now, the claim I'm making is: sentence meaning radically underdetermines the content of what is said. (Searle 1992, p. 181)
Thus, Searle explains, nothing in the literal meaning of the sentence referred to excludes crazy interpretations like: “John opened the door with the key by swallowing both door and key, and moving the key into the lock by way of the peristaltic contraction of his gut.” (Searle 1992, p. 182) Note that we are dealing with a claim about linguistic meaning here, not about semantic content—properly construed as representational content, uniquely determining satisfaction conditions.
From the viewpoint of the social-epistemological picture of semantic content sketched above, the Background Hypothesis should be restricted to the producers of sentences figuring in linguistic representation. On this picture, only the producers' intentionality requires background know-how regarding the application of those sentences. Mere consumers merely need an appropriate background of social practices. If the advocate of this picture did not restrict the Background Hypothesis to the producers, he would be committed to the view that mere consumers can give voice, in the narrow sense, to intentional states in which they cannot be, due to lack of background know-how. This would mean that only the producers can be sincere in their assertive utterances of sentences regarding which they are producers. But this seems wrong. It is possible for mere consumers to deliberately express knowledge by testimony. Hence, the (unrestricted) hypothesis of the Background ought to be rejected, on the present view. Meaning-intentions (meaning-bestowing acts) do not generally require a non-intentional background relative to which their (truth-conditional) content and satisfaction conditions are determined; while their intuitive fulfilments (the corresponding “verifications”), if any, do. For instance, it is impossible to perceive something as an elm without being able to distinguish elms from other sorts of trees. This of course means in turn that one ought to reject the present picture if one accepts the Background Hypothesis (in unrestricted form). In order to decide the issue, more needs to be said to explain this hypothesis. I cannot decide the issue here. But it may be helpful in this regard to end by saying a bit more about the content of Searle's Background Hypothesis.
In The Rediscovery of the Mind Searle plausibly contends that mental representation, i.e., underived, original intentionality is realized just in case a given mental state “is at least potentially conscious” (Searle 1992, p. 132). We find similar claims in Husserl.[18] Due to the “aspectual shape” of intentional states (the fact that they have perspectival, intentional content) there are no “deep unconscious mental intentional phenomena” (Searle 1992, p. 173), such as reflectively inaccessible belief states. There is an important sort of background elements whose distinctive mark is that they are capacities to be in intentional states; that is, they are dispositions to have (actually or potentially) conscious representations, such as occurrent beliefs. The general assumption that things offer resistance to pressure is a case in point. We normally do not form a belief to this effect but are nevertheless committed to it by the way we behave towards things (cf. Searle 1992, p. 185).
One may call these capacities for (at least potentially) conscious representation “background assumptions” or “network beliefs” if one likes, but according to Searle one must keep in mind that these capacities fail to be intentional states: “the Network of unconscious intentionality is part of the Background” (Searle 1992, p. 188) and “the Background is not itself intentional” (Searle 1992, p. 196). If Searle is right about this, then many elements of the so-called “web of belief” are part of the non-intentional background.
This view has far-reaching consequences for the theory of intentionality. For, if Husserl is basically right about the structure of consciousness (as I believe he is), then conscious states must be embedded in a holistic structure, which Husserl calls the “intentional horizon,” whose future elements are predelineated (at least in part) by the intentional content of the respective state of consciousness. For example, if you consciously see something whose front side you are visually confronted by as a house, then you will anticipate[19] visual appearances of a back side and an inside, respectively, as future experiences you would undergo if you walked inside or walked around the object while observing it. But is the corresponding set of anticipations really an intentional structure? Searle’s arguments regarding the background cast doubt on this, given his view that consciousness (or what is consciously accessible) is the only occurrent reality of intentionality. After all, it is plausible to equate (a large subset of) the set of anticipations determining the respective intentional horizon with a relevant part of what Searle calls the “Network,” given that they cannot be described properly as occurrent beliefs or conscious judgments, but rather as mere dispositions to form higher-order beliefs. For, as Husserl explains, the anticipations in question concern the way the represented object would present itself to consciousness in possible worlds compatible with what is currently experienced, and they also concern the way this object relates to other objects in the world—thus constituting the core of one’s current world horizon, which core Husserl calls the “external horizon” (Husserl 1973, p. 32) of the experience (see below). It is only when these anticipations are intuitively fulfilled, in the sense that relevant conscious episodes of (what seem like) verification (such as perceptual verification) occur, motivating corresponding acts of judgment, that there will be entries into the relevant mental dossier associated with the object in question. As Husserl puts it (referring to mental dossiers associated with proper names as “individual notions”):
I see an object without an ‘historic’ horizon [footnote: without a horizon of acquaintance and knowledge], and now it gets one. I have experienced the object multifariously, I have made ‘multifarious’ judgements about it and have gained multifarious [pieces of] knowledge about it, at various times, all of which I have connected. Thanks to this connection I now possess a ‘notion’ of the object, an individual notion […] [W]hat is posited in memory under a certain sense gains an epistemic enrichment of sense, i.e., the x of the sense is determined further in an empirical way.[20] (Husserl 2005, p. 358; my translation)
The “historic” horizon and the objects of the relevant anticipations constitute the “internal horizon” (Husserl 1973, p. 32) of the experience. They all belong to the same “x of the sense” (also referred to by Husserl as the “determinable X”), i.e., they share a sense of identity (of represented object) through time. Other past and anticipated experiences bring it about that one’s “‘notion’ of the object” is networked with other notions of objects. They constitute the external horizon of the experience.
If the anticipations in question were part of a non-intentional background, then it would be wrong, of course, to describe them as being directed at objects; as a consequence, the Husserlian conception of intentional horizon just sketched would break down. To avoid this consequence, Searle's Background conception needs to be altered, such that the background may indeed contain intentional elements, albeit in a derived sense.
This can be fleshed out as follows. The primary bearers of intentionality are (at least potentially) conscious units, such as judgments and the experiences that motivate them. It is true that respective meaning and intentional content only function against a background the elements of which lack this primary form of intentionality. However, this background contains some elements that possess a derived form of intentionality, so that it is misleading to describe it as completely non-intentional. In particular, it contains mental capacities or dispositions to form beliefs about the further course of experience which Husserl (in 1973, para. 8) calls “anticipations.” Some of the experiences thus anticipated correlate with an internal horizon. Their occurrence may lead to entries being made in a mental dossier, which are empirical beliefs (informational states) to which a “referent” (an object they are about) can be assigned in a principled way, in accordance with the modified principle of knowledge maximization. Here is an example of such a principle of reference assignment, which I have proposed in earlier work.[21]
The logical subject x of […] a belief of the form a is F […] whose acquisition goes together with the opening of a mental dossier about x is identical with the logical subject y of the judgement initiating that belief (or x would be identical with y, if x and y existed). (Beyer 2001, p. 287)
“Logical subject” here refers to the object the relevant belief is about (such as table 1 in the case of the persisting belief actualized by the judgment given voice to at t2 in the above example about the wobbling table); and the judgment initiating that belief is understood to have its logical subject assigned in accordance with the modified principle of knowledge maximization, as explained at the end of section 3, above.
I conclude, first, that the background of meaning and intentional content may be looked upon as being at least in part itself intentional, albeit in a derived sense, but that, second, the applicability of the Background Hypothesis still needs to be restricted, as far as the part of the background (co-)determining truth-conditional content is concerned, to what I have called the producers.