We cannot say that theories are definitely true or completely supported by evidence, however (1993: 203-222). Justifying theories is rather a matter of probability or graduality;
“justification is not categorical, but comes in degrees” (op.cit.: 222):
Not all scientific claims are either accepted as definitely true or rejected as definitely false, nor should they be; evidence may be better or worse, warrant stronger or weaker, and the acceptance status of the claim can, and should, vary accordingly (1998: 110).
This is so because both “our theories about the world and ourselves” as well as “our criteria of justification” are fallible (1993: 222):
[…] we can have no proof that our [Haack’s] criteria of justification are truth-guaranteeing, but reasons for thinking that, if any indication is available to us, they are truth-indicative; reasons no less fallible than those parts of our theories about the world and ourselves with which they interlock, but no more so, either (ibid.).
The fallibility of our theories and criteria of justification is due to our “real, imperfect”
constitution as human beings (1998: 97):
[…] any actual scientific community consists of real, imperfect human beings, […] individual idiosyncrasies or weaknesses may [however] compensate for each other. […] in a community of inquirers, some will be more conservative in temperament, inclined to try adapting an old theory to new evidence, others more radical, readier to look for a new approach. I doubt that real scientists are ever quite single-mindedly devoted to the truth […]. But to the extent that science is organized so […] that partisans of one approach seek out the weaknesses which partisans of another are motivated to neglect, a real community of imperfect inquirers can be a tolerable ersatz of an ideal community (op.cit.:97-98).
Real communities of inquirers are always imperfect, although potentially less imperfect than an individual inquirer working on her own without correction from other inquirers. The most we can hope and work for are real communities of inquirers which are organized in ways that make them able to approach what we would consider to be the ideal epistemic community, and so produce theories that are as truth-indicative as possible, even if their truth can never be guaranteed. But even our ideal “hypothetical” notion of an epistemic community, even the best criteria of justification we are able to articulate, are fallible (1993: 214). That is to say:
Not only is complete justification of theories faithful to our best criteria of justification impossible; if complete justification of theories was in fact possible, this would not guarantee
that the theories were true, because the best criteria of justification imaginable by real, imperfect, fallible human beings are themselves fallible.
The best criteria of justification imaginable under the condition of fallibilism are what Haack refers to as the “foundherentist”23 criteria of justification; justification faithful to considerations of “supportiveness”, of “independent security” and of “comprehensiveness”
(op.cit.: 73, 87). Genuine inquiry; inquiry where these considerations are taken properly into account, are compared with a crossword puzzle: “The structure of evidence” is regarded […] as analogous to a crossword puzzle according to which an empirical proposition is more or less warranted depending on how well it is supported by experiential evidence and background beliefs (analogue: how well a crossword entry is supported by its clue and other completed entries); how secure the relevant background beliefs are, independently of the proposition in question (analogue: how reasonable those other entries are, independent of this one); and how much of the relevant evidence the evidence includes (analogue: how much of the crossword has been completed) (1998: 105-106).24
Thus, whether a proposition should be included in a theory or not, depend on how well it is supported by independently secure and comprehensive evidence, i.e. on the degree of “its explanatory integration”25 (ibid.). Evidence, moreover, is “personal rather than impersonal”
(1993: 20):
The explicandum is: A is more/less justified […] in believing that p, depending on … . The choice of explicandum […] indicates […] that it is a personal locution, not an impersonal locution like ‘the belief that p is justified’ (op.cit.: 73).
Haack defends an “epistemology with a knowing subject” (op.cit.: 97-98); the propositions of theories are more or less warranted for someone.26 Epistemology is precisely a matter of
23 Haack describes foundherentism as the outcome of a critical reconstruction of foundationalism and coherentism. Coherentists subscribe to the thesis that a belief is justified if it belongs to a coherent set of beliefs.
Foundationalism is based upon two theses; i) that some justified beliefs are basic; a justified belief is justified not by the support of any other belief, but by the subject’s experience, and ii) that all other justified beliefs are derived; a derived belief is justified via the support, direct or indirect, of a basic belief or beliefs.
Foundherentists, like Haack, claim however i) that a subject’s experience is relevant to the justification of empirical beliefs, but that there need be no privileged class of empirical beliefs justified exclusively by the support of experience, independently of the support of other beliefs, and ii) that justification is not exclusively one-directional, but involves pervasive relations of mutual support (1993:10-33).
24 For a detailed and systematic outline of “the crossword puzzle of inquiry”, see Haack (1993: 81-89).
25 “[…] by appealing to the notion of explanatory integration in the explication of supportiveness, foundherentism borrows some of the intuitive appeal of the notions of (on the foundationalist side) inference to the best explanation and (on the coherentist side) explanatory coherence” (1993: 84).
26 The outline of an epistemology with a knowing subject is done in explicit opposition to Karl Popper’s
“championship of an epistemology without a knowing subject concerned solely […] with propositions and their logical relations” (1993: 101).
explicating the best criteria of justification for a community of inquirers. Referring to Charles Sanders Peirce, Haack considers the ideal epistemic community, the community imaginable most likely to produce truth-indicative theories, to be a community where investigations, faithful to the foundherentist criteria of justification, continue indefinitely. Peirce characterizes truth as
[…] the ultimate representation, the Final Opinion, compatible with all possible experiential evidence and the fullest logical scrutiny, which would be agreed by all who investigate were inquiry to continue indefinitely (1998: 162).
Or as Peirce himself puts it: Truth is “the opinion that would be ultimately agreed by all who investigate”, as “that concordance of an abstract statement with the ideal limit towards which endless investigation would tend to bring scientific belief” (quoted in op.cit.: 166, n. 25).
Peirce did not defend an anti-realist, purely “conversational”27 consensus theory of truth and justification, however. If he had, Haack could not have made him an ally: Anti-realism is incompatible with foundherentism, constructed not to be vulnerable to what she refers to as
“the drunken sailor’s argument” so “fatal to coherentism” (1993: 27):
The fundamental objection is this: that because coherentism allows no non-belief input – no role to experience or the world – it cannot be satisfactory; that unless it is acknowledged that the justification of an empirical belief requires such input, it could not be supposed that a belief’s being justified could be an indication of its truth, of its correctly representing how the world is (ibid.).
With Peirce, Haack considers “the real” both as independent of what “you or I or anybody”
thinks it to be (1998: 163) – there is a “world” from which we can get “non-belief input”
(1993: 27) – and as what is “ultimately” represented in “the Final Opinion” of the ideal community of inquirers (1998: 162). There are two presuppositions for this “reconciliation”
(op.cit.: 163). One is the interpretation of reality as something within the reach of “possible cognition” (ibid.). Haack regards talk about “absolutely incognizable” reality as
“pragmatically meaningless”: “it is impossible to make sense of any question to which we could not, however long the inquiry continued, determine the answer” (ibid.). This peculiar
“repudiation of a world of unknowable things-in-themselves” is also defended by Peirce.
Haack quotes Peirce saying:
27 Haack dismisses Rorty’s “vulgar pragmatist” conception of justification as “conversation” (1993: 182-202).
[…] the highest concept which can be reached by abstractions from judgments of experience – and therefore, the highest concept which can be reached at all – is the concept of something of the nature of a cognition […] Not, then, […] is a concept of the cognizable. Hence, not-cognizable […] is, at least, self-contradictory […]. In short, cognizability (in its widest sense) and being are not merely metaphysically the same, but are synonymous terms (original emphasis, quoted in ibid.).
It is this move, the equation between being and cognizability, between the world and the world-for-us, that enables Peirce (and Haack) “to avoid the hopeless obsession with the skeptical challenge to which more rigid realisms seem drawn” (op.cit.: 164).
The second presupposition for the reconciliation of the real and the Final Opinion is that non-belief input from the world, entering the human mind through perception, is considered epistemologically relevant. Our capacity to perceive and to let our beliefs be informed by what we perceive, i.e. by our sense experiences, is regarded as a cognitively significant capacity – we may be able to know better, to construct theories that are better justified, because of it. This is to allow for causal relations in justification, and a moderate naturalism in epistemology, since human perception is a process that can be studied empirically and elaborated causally.28 However, this does not allow for epistemology to be completely naturalized; to say “that traditional epistemological problems […] are illegitimate [problems], and should be replaced by new natural-scientific projects” (1993: 4). Justification should, rather, be considered an “interplay of causal and evaluative aspects” (op.cit.: 73). It involves, unavoidably, “normative” considerations (1998: 108): Sense experiences are epistemologically relevant in the sense that they are made epistemologically relevant by inquirers who let their investigations be directed by the best criteria of justification available (i.e. the truth-indicative criteria of foundherentism).
This notion of justification highlights, moreover, the intimate connection between perception and conceptualization in inquiry. The fact that “our perceptual judgments are conceptualized, interpretative” should, however, Haack notes, not allow us to forget that “in perception we are in contact with something real, independent of our interpretations, of how anybody thinks it to be”: Perception involves “a potential for surprise” (1998: 161-162). We cannot conceptualize reality – which is what we try to do when we pursue genuine inquiry – without
28 Again, this point is directed especially against Karl Popper’s account that considers “only relations of deductive logic” as “epistemologically relevant” (1993: 101): Causal relations, and thus “scientists’ perceptual experiences”, can play no role in justification, according to Popper (op.cit.: 99).
conceptualizing. Nor do attempts to conceptualize reality make sense if they are not in fact attempts to conceptualize (cognizable) reality. There are many ways of conceptualizing the surprises of our encounters with the real: “There are many different vocabularies, and many different true descriptions of the world” (op.cit.: 157). Haack even stresses “that cognitive advance is not always a matter of new claims in an old vocabulary, but often a matter of conceptual innovations marked by new vocabulary, or by shifts in the meaning of old vocabulary” (op.cit.: 160). She insists, however, that
[…] if there is complete failure of translatability [between descriptions], there is compatibility. If, on the other hand, there is translatability, there may be compatibility or incompatibility. If the different descriptions are incompatible, they cannot both be true […].
But if the different descriptions are compatible […] the different true descriptions can be conjoined in a single (even if heterogeneous) true description (op.cit.: 161).
Thus, there is no “real incommensurability” (op.cit.: 96). Propositions of different vocabularies, however, can in the end be judged compatible or incompatible with other propositions: A proposition in any language whatsoever is warranted if it contributes sufficiently to the explanatory integration of a theory. Any proposed theory, however conceptualized, is justified if it is supported by independently secure and comprehensive evidence. The foundherentist criteria of justification are considered generally valid, i.e. as the proper standard of epistemological assessment of propositions of all vocabularies. Hence, with her construction of the crossword puzzle model of inquiry, Haack denies notions of
“deep”, “normative or philosophical” “epistemic” and “ontological” relativism (op.cit.:
148).29 For example is the problem with Richard Rorty’s notion of justification as conversation, both its anti-realist presuppositions, and its “relativist and cynical”,
“contextualist + conventionalist” approach to epistemology (1993: 193).30 The foundherentist
29 Haack mentions also “normative or philosophical” “moral” and “aesthetic” relativism, but she does not address these relativisms (1998: 148). In her analysis of epistemic and ontological relativism, she introduces a set of finer distinctions: Both “epistemic value”, “meaning”, “reference”, “truth”, “metaphysical commitment”,
“ontology” and “reality” have been presented – by different people – as being relative to “culture or community”, “language”, “conceptual scheme”, “theory”, “scientific paradigm”, “version, depiction, description” – or even to the “individual” (“subjectivism”) (op.cit.: 149-166). The deep, normative or philosophical relativism – which in Haack’s view is false – is positioned in opposition to “shallow”, “descriptive or anthropological” relativism “to the effect that different communities or cultures accept different epistemic (moral, or aesthetic) values” – which might be true, but which is philosophically uninteresting (op.cit.: 151).
30 Haack notes, however, that Rorty has lately defended “tribalism + conventionalism”, and not really
“contextualism + conventionalism”. Tribalism is not relativist, because it claims that “A is justified in believing that p if A satisfies the criteria of our epistemic community”; it is an ethnocentrist position. The tribalist move however does “not get him [Rorty] off the hook […]. Tribalism is entirely arbitrary and unmotivated unless one thinks that the criteria of one’s own epistemic community are better than those of other communities; that is, it
criteria of justification are considered, rather, to be the best standard of warrant available in any context (under the condition of fallibilism), because they are the standard most likely to produce theories that are truth-indicative.
The question is why truth is so important. Why should justified theories indicate truth? Why not prescribe justified beliefs as the aim of inquiry, without presupposing that justified beliefs are (most probably) true? Rorty states that “it makes no difference in practice whether you aim at the truth, or aim at justified belief” (1998b: 20). Haack maintains, however, that “to believe that p is to accept p as true” (1993: 192); “truth is the internal goal of belief” (1998:
16). She does not consider this to be “a sophisticated remark about truth” (1993: 192). It is rather a “truism about belief” (ibid.). It is “a tautology” to argue that inquiry, where we scrutinize our beliefs, aims at the truth (1998: 189): “If you aren’t trying to find out the truth about whatever-it-is, you aren’t really inquiring” (ibid.). But why bother really inquiring?
Why seriously believe anything? Why “engage fully – non-cynically” in genuine investigations (1993: 192)? What is so upsetting about the “fake” and “sham” reasoning of pseudo-inquiry, i.e. with being either indifferent to the truth-value of a proposition for which one seeks to makes a case (fake reasoning), or not wanting to discover the truth of some question, but to make the case for some proposition to which one has a “prior and unbudgeable commitment” (sham reasoning) (1998: 9)? Generally, because non-cynical inquiry is instrumentally valuable in a very crucial sense: Haack is convinced that genuine truth-seeking serves the survival of the human species (op.cit.: 13-14).31 In addition, genuine inquirers are morally virtuous: To avoid fake and sham reasoning is a matter of being a
“decent”32 academic and a “good”33 person.
pulls against conventionalism, to which, however, Rorty is unambiguously committed” (original emphasis, 1993: 192-193).
31 “Compared with other animals, we humans are not especially fleet or strong; our forte is a capacity to figure things out, and hence to anticipate and avoid danger. Granted, this is by no means an unmixed blessing; as shrewd old Thomas Hobbes put it long ago, the same capacity that enables men, unlike brutes, to engage in ratiocination, also enables men, unlike brutes, ‘to multiply one untruth by another’ […]. But who could doubt that our capacity to reason – imperfect as it is, and easily abused – is of instrumental value to us humans?”
(1998: 13-14).
32 “It seems almost indecent when an academic whose job is to inquire, denies the intelligibility or denigrates the desirability of the ideal of honest inquiry” (1998: 14). The indecency stems not from the often harmful consequences of what Haack refers to as over-belief (believing beyond what one’s evidence warrants) or under-belief (not believing when one’s evidence warrants under-belief) as such, but rather from the fact that the inquirer can be held responsible for these consequences: It is because the damage “results from self-deception, from a lack of intellectual integrity, that it is morally culpable” (op.cit.: 15).
33 “To be sure, intellectual integrity is not sufficient by itself, any more than courage is, or kindness, to make you a good person […]. And, yes, you might be in other respects a decent person […], while lacking in intellectual honesty. But, to my ear at least, ‘he is a good man but intellectually dishonest’, if not an oxymoron, really does not need an ‘otherwise’” (1998: 15).