TESTIMONY AND INTELLECTUAL VIRTUES IN HUME’S EPISTEMOLOGY
Ruth M. Espinosa[1]
Abstract: In this paper, I consider some issues concerning Hume’s epistemology of testimony. I’ll particularly focus on the accusation of reductivism and individualism brought by scholars against Hume’s view on testimonial evidence, based on the tenth section of his An enquiry concerning human understanding. I first explain the arguments against Hume’s position, and address some replies in the literature in order to offer an alternative interpretation concerning the way such a defense should go. My strategy is closely connected with Hume’s notion of virtue and the role it plays in his epistemology, mainly as presented in his A treatise of human nature. I address the problem of how the section “Of miracles” in the Enquiry must be properly understood, as several misunderstandings of Hume’s epistemology of testimony emerge partially from the particular character and aim of that section.
Keywords: David Hume. Testimony. Miracles. Virtue. Epistemology.
1 Testimony, reductive thesis, and virtue
1.1 Reductive thesis (RT)
There is a long-standing debate in contemporary social epistemology concerning the status of testimonial grounding for belief and knowledge. Hume's own treatment of that topic, sometimes recognized as the “the received view” (COADY, 1992, p. 79), has stood in the middle of such polemic since his account has been considered by leading contemporary scholars as reductive and individualist. Thus, for example, G. E. M. Anscombe (1979, p. 143) claims:
The greater part of our knowledge of reality rest upon the belief that we repose in things we have been taught or told. Hume thought that the idea of cause and effect was the bridge enabling us to reach any idea of a world beyond personal experience. He wanted to subsume belief in testimony under belief in causes and effects, or at least to class them together as examples of the same form of belief [...] It was always absurd, and the mystery is how Hume could ever have entertained it.
More or less along the same lines, C. A. J. Coady[2] has also assessed Hume's account of testimony as insufficient and misleading, for his theory consists of a “reduction of testimony as a form of evidence or support to the status of inductive inference” (COADY, 1992, p. 79). Inductive inference, in turn, is due to regular conjunction of experienced events.
We are told by Hume that we only trust in testimony because experience has shown it to be reliable but where experience means individual observation and the expectations it gives rise to, this seems plainly false. (COADY, 1992, p. 80).
Coady’s and Anscombe’s statements against Hume’s reductionism is a consequence of a deeper concern, namely, that subsuming testimony - either its nature or its epistemic quality - to firsthand experience, speaks against a proper understanding of the social dimension of knowledge. Human beings are epistemic dependent[3] to a great deal and the question that stands is how epistemology now and then, is to be sensitive to this dependence, especially while dealing with testimonial belief and knowledge. C. A. J. Coady is one of the first scholars who brought attention to this topic when most epistemologists were tacitly subsuming testimony under other sorts of evidence like memory or perception. Coady brings to our attention also that reductionism seems to be problematic because it leads to individualism.
Now, reductionism about testimony seems to be incompatible with non-individualist epistemology, and this is exactly what Coady observed to be a major flaw in Hume’s account:
We find Hume speaking of “our observation of the veracity of human testimony” and ‘our experience of their constant conjunction [...] Then R.T. (reductionist thesis), as actually argued by Hume, is involved in vicious circularity since the experience upon which our reliance upon testimony as a form of evidence is supposed to rest is itself reliant upon testimony which cannot itself be reduced in the same way. The idea of taking seriously someone's else's observations, someone else's experience, already requires us to take their testimony (in this case reports of what they observe) equally seriously. (COADY, 1973, p. 240-241).
This last reason, Coady thinks, suffices “on Hume's behalf” to “retract his incautious commitment to common experience” (COADY, 1973, p. 241).
As it can be noted in the former quote, Coady rejects the claim that testimony can be reduced to first-hand non-testimonial evidence, since it rules out the possibility to account for common experience or common ground for belief and knowledge without incurring in vicious circularity. Given our epistemic dependence, that is to say, the impossibility to personally obtain the evidence needed to support all our beliefs, and the fact that reductionism would lead either to circularity, or into a dramatic narrowing of the scope of human knowledge to first-hand individual evidence, Coady raises the powerful claim that testimony is a distinct sort of evidence[4], which differentiates it from other sources of knowledge, like empirical evidence, memory or probability, that it is unsolvable related to social practice. Testimony is thus taken to be a certain speech act, namely an illocutionary act “which may be and standardly is performed under certain conditions and with certain intentions” (COADY, 1992 p. 25). As such, testimony may be understood as someone’s expression of belief. The assertion that expresses belief, hides a number of relevant features of that belief, such as its factual support, the degree of confidence, and, most importantly, its sincerity[5]. Thus, while belief can be true or false, testimony can be also sincere or insincere. Other sorts of evidence are not intentional and have not this peculiar character. Nonetheless, the question remains, whether this shapes testimony and its epistemic assessment differently.
In conclusion, Coady’s criticizes Hume’s epistemology of testimony, for testimony would have been treated as any other kind of evidential support for belief. As such, its reliability would depend on a correlation of facts (testimonial reports) as experienced in the past. Moreover, since experience is understood ad individual observation, reductionism involves individualism, what in turn, oversees what is distinctive about the ‘act of testifying’, whether it be “formal” (as in a law court), or “natural”[6] as it is involved in everyday life. As to the specific nature of testimony, Coady points out two general arguments against the reductionist thesis. First, Coady claims that it is absurd “to suggest that, individually, we have done anything like the amount of fieldwork that RT requires” (COADY, 1973, p. 246), in other words, we don’t have the appropriate amount of first-hand evidence to judge others’ reports in most situations. For the same reason, an appeal to empirical generalization of the type “kinds of situations” or “kinds of reports” by Hume, is equally ungrounded, for the ability of expertise or authority to ground someone´s testimony as reliable, depends on the possibility of establishing some kind of empirical correlation between reports and reality at the particular level, or by relying on testimony itself.
Secondly, it would follow from Hume’s reductionism that the relationship between testimony and reality is only contingent, subject to evaluation case by case. Coady argues that such a premise would speak against the possibility of a public language. Coady explains himself through the following thought experiment. Imagine a community of Martians who never make true utterances, that is, a context in which no report holds a relationship with reality, in such a context a “humean correlation” couldn't possibly be verified (say, by learners of that language), but “the ability to make true reports with words is connected with using the words correctly and this ability is something that can only be exhibited in the consistent making of true reports” (COADY, 1973, p. 154). Thus, the very possibility of learning and the understanding of a language requires the existence of at least one true report. And this counts as an a priori reason to trust testimony. A core requirement for a proper account of testimony has now become clear. It seems that testimony must be recognized as having a primitive status as evidence for belief, otherwise, a wide array of social epistemic structures, cannot be accounted for without begging the question. Thus, relying on testimony cannot be justified by reducing it to perception or any kind of inference whose sole possibility already presupposes it.
1.2 Responses to Coady’s criticism
Coady's reconstruction of Hume as a paradigmatic case of reductionism is at odds with many intuitions of Hume's philosophy. Speaking generally, his reconstruction oversimplifies Hume’s understanding of experience. Thus, for example, it has been pointed out that the kind of inference by which we – on the basis of past experience – according to Hume, “place credit in witnesses or historians” (EHU 10, SBN 113) doesn't have to be understood as a case of “conscious explicit reasoning” (WELBOURNE, 2002, p. 412). According to Michael Welbourne, especially in cases in which the basis of the report is totally unknown to us, it seems that the kind of association that takes place in belief formation is one between testimony 'as such' and reality. Moreover, he suggests that Hume shows in some passages to have grasped the intuition that the kind of speech-act performed in testimony is intrinsically associated with the kind of response intended by the speaker, that is, acceptance of the facts as a default response (cf. WELBOURNE, 2002, p. 416). This seems to be the case when Hume writes:
If one person sits down to read a book as a romance, and another as a true history, they plainly receive the same ideas, and in the same order; nor does the incredulity of the one, and the belief of the other, hinder them from putting the very same sense upon their author. (THN 1.3.7.6 SNB 97-98).
Thus, learning in the practice of “informative telling” (testimony-like speech-acts) and developing the proper answer are not “two different lessons”. Welbourne emphasizes that for Hume “we learn the act and the looked-for response as a single package just as we do with ordering and obeying. Hume’s modest claim, for which he offers little argument – though perhaps not much is needed – is that we learn these things by being exposed to exercises of the practice.” (WELBOURNE, 2002, p. 418). Learning from other’s reports is, indeed, in line with belief formation from inductive inference, as it is based on the expectancy that constant conjunction gives rise to. Nevertheless, whereas in the latter the natural response of the mind when placed in the characteristic circumstances is belief, in the former case is acceptance based on trust (see EHU 5.1.4-5, SBN 42-3), the mechanics behind is shared by testimonial and non-testimonial evidence, and both are by no means reducible to individual observation. Testimony as such, is then among the building blocks of historical knowledge, whose possibility rests on two conditions, namely, our ability to reason causally, that is, “beyond our senses”[7], and the constancy of human nature[8]. Whether his theory of historical knowledge is accurate or not, the important thing is that Hume’s conception of experience cannot be reduced to individual observation precisely because of the importance of such topics in his account of belief. When testimony takes place within our reasoning, social elements concur along with facts. Other’s beliefs, on the other hand, must be epistemically assessed, just as any causal reasoning should:
When we receive any matter of fact upon human testimony, our faith arises from the very same origin as our inferences from causes to effects, and from effects to causes; nor there’s anything but our experience of the governing principles of human nature, which can give as any assurance of the veracity of men. (THN 1.3.9.12 SBN 113).
Axel Gelfert has developed a related point concerning Coady’s criticism. He notices that Coady’s reconstruction of Hume on testimony focuses on Of miracles[9]. As much as that piece contains Hume’s most extensive treatment of testimony, its assertions must be considered contextually. Gelert, thus, suggests that it could be problematic for Coady’s stand to focus on a text devoted to discussing testimonial basis for miracles because it risks distorting “epistemic merit of testimony considered in its own right, i.e. as a sort of probable knowledge” (GELFERT, 2009, p. 67). Indeed, the core argument in Of miracles seeks to discredit miraculous testimony for reasons that are internal to that specific kind of testimony, namely, its behavior towards laws of nature. Hume defines ‘miracle’ as a “violation of the laws of nature”, which in turn, are established by a “firm and unalterable experience” (EHU 10.1.12 SBN 114). This sort of uniform experience is what constitutes a proof in his jargon. That is why Hume sustains that “no testimony is sufficient to establish a miracle” (EHU 10.1.12 SBN 116). Testimony that intends to accredit that a violation of a law of nature has indeed occurred is something that simply overruns what the 'institution of testimony' can do. A kind of testimony, which is necessarily unreliable, can hardly be considered as such, precisely because, as Hume suggests, rationality is a matter of responsiveness to reasons. The wise, Hume claims “proportion his belief to evidence” (EHU 10.1.4 SBN 110). Gelfert’s proposal, is very sensitive to this fact, he writes that “Hume merely demands that no exception be made in favor of testimony, when it comes to drawing inferences on its basis; in this respect, testimony in entirely on a pair with direct personal experience, in that inferences must be founded ‘on our experience of their constant and regular conjunction’” (GELFERT, 2009, p. 66). This is indeed enforced in Hume´s Treatise as well as in his Enquiry and shows the way he addressed the difference between the nature of testimony, and its epistemic evaluation:
The words or discourses of others have an intimate connection with certain ideas in their mind; and these ideas have also a connection with the facts or objects which they represent. This latter connexion is generally much over-rated, and commands our assent beyond what experience will justify. (THN 1.3.9.12 SBN 113).
If Hume is ultimately not committed to a reductionism (RT) concerning testimony, it seems that Coady's argument for the self-contradictory character of invoking common experience also vanishes. Under these circumstances, we still need to raise the question of how and under which conditions epistemic agents are justified in believing others’ reports, in what measure it depends on knowledge of human nature and whether those conditions are subject-dependent or maybe also community-dependent.
1.3 Relying on testimony: a matter of virtue?
Michael Root has advanced an interesting point of view concerning the relation between testimony and virtue. Root departs from a reconstruction of the conditions for the reliability of testimony in Of miracles:
“When A testifies that p to B, B should estimate
(1) the prior probability that not p, and
(2) the credibility of A as a witness that p (the unlikelihood that A is lying or mistaken).”
The relevant question is then: “On what can B base his estimate of (2).” (ROOT, 2001, p.19).[10]
He proposes a complementary escape route of reductivism. Instead of assuming that B can only base his or her estimation of the reliability of A's belief on the basis of his or her past experience that witnesses like A have most of the time, spoke the truth, Root’s proposal goes beyond explicit textual evidence in suggesting to apply Hume's distinction between natural and artificial virtue to the present issue. Veracity can, thus, be considered an artificial virtue in a certain community whose members share a common end, namely, to gain knowledge. Root (2001, p. 21) claims: “If giving honest or well-founded testimony is an artificial virtue, then B has a reason to believe that A is a credible witness, a reason to expect that, in testifying that p, A is conforming her testimony to the truth that is based on convention rather than in past conformity.”
What is especially interesting about this argument is that, on the one hand, it changes the relation between testimony and truth from one based on past experience, to a conventional one – the reason we expect a testimony to be true is because we “count with” the fact that the epistemic agents in such a community have the right motivation not to lie and to make her best effort to avoid error. On the other hand, B would also have at her disposition “forward-looking” reasons for believing that A will speak the truth. This is not opposed, however, to Hume’s idea that epistemic assessment of testimony requires knowledge of human nature. As to the particular circumstances of the scientific community, it is the common motivation to find truth what makes it safer for us to judge it as reliable. The assessment of testimony under these circumstances does not depend on past correlations between testimony and reported facts.
Root’s proposal is oriented towards the cooperative scheme of a scientific community. He writes:
The virtue of testimony, honesty and respect for the evidence enable members of a scientific community to base their beliefs on the testimony of others without looking to past conjunctions between their testimony and the truth. Members are not born to tell the truth, any more than they are to keep their promises. Telling the truth and keeping promises are both human contrivances, schemes of cooperation based on mutual intention and sustained by self-interest and moral approbation. (ROOT, 2001, p. 30).
This interpretation rightly connects the issues of virtue and testimony in a way that allows us to integrate Hume's conception of the role of testimony in gathering knowledge and forming true beliefs while avoiding reductivism. The constitution and existence of a scientific community presuppose the responsible and rightly-motivated epistemic agents, but furthermore, it presupposes the possibility to explain such rational behavior within the science of human natures. Reliability of testimony as such is subdued to an assumed intimate commitment of practitioners to veracity, thus involving virtue in testimonial exchange within a community. However, there still is a limitation of trust to experience, as well as in miraculous reports, for vices as credulity must be condemned since it “commands our assent beyond what experience will justify”. Deception, laziness in research, insincerity, lack of rigors, represent also a hazard to the scientific enterprise itself, but not only.
To my mind, Roots’ view faces two problems. On the one side, it lacks textual evidence. Though there is much evidence for the social character of knowledge in Hume's writings, there is scarcely any evidence for placing scientific veracity, within the framework of artificial virtue. It might work well systematically, but I think Hume just didn't need that resource. As I have claimed, Hume’s way to epistemic assessment of testimony is through the knowledge of human nature, its passion, motivations, and natural vice and virtue. On the other side, the soundness of belief based on the presupposition of veracity and reliability of testimony, when understood as an artificial virtue of the scientific community, is limited to the context of that community only. I will argue in the next section for a broader account.
2 Hume's normative epistemology and virtue
2.1 Testimony as such
Let’s start with the following passage:
Were not the memory tenacious to a certain degree; had not men commonly an inclination to truth and a principle of probity; were they not sensible to shame, when detected in a falsehood: Were not this, I say, discovered by experience to be qualities inherent in Human nature, we should never repose the last confidence in human testimony. A man delirious, or noted for falsehood or villany, has no manner of authority with us. (EHU 10.1.4, SBN 112).
This passage is core to this proposal, since, I claim, it shows the kind of connection between testimony, epistemic virtue, and truth, that is overseen in Roots and the other approaches. A closer consideration of the passage will show that the problem of reductivism can be solved at a much more fundamental stage of the theory. I will argue that (1) empirical correlations to judge the credibility of particular reports can be seen as independent of testimony as such, and (2) virtue is needed at the stage of estimating the reliability of belief in general.
(1) According to the passage, the reason we place confidence in other’s words is twofold: on the one hand, there’s a natural inclination to truth, on the other, there’s fear to be found untrustworthy. Both are explained by Hume in our natural self-interest and sympathy, which lead us to care for our own good as much as for the opinions others hold about ourselves. These natural tendencies have already come out in the discussion above since Hume claims that is nothing but “our experience of the governing principles of human nature, which can give us any assurance of the veracity of men” (vid. supra).
The very same conceptual resources explain a great deal of conventional normative structures in society, such as morality or law. A convention arises because of natural forces, and does not even require an explicit commitment to it, there is a general sense of common interest which explains it:
Two men, who pull the oars of a boat, do it by an agreement or convention, tho’ they have never given promises to each other. Nor is the rule concerning the stability of possession the less deriv’d from human conventions, that it arises gradually, and acquires force by a slow progression, and by our repeated experience of the inconveniences of transgressing it. On the contrary, this experience assures us still more, that the sense of interest has become common to all our fellows, and gives us a confidence of the future regularity of their conduct: And ’tis only on the expectation of this, that our moderation and abstinence are founded. In like manner are languages gradually establish’d by human conventions without any promise. (THN 3.2.2.10 SBN 490).
Hume’s example, allows a few remarks on testimony, even though, testimony is not the topic in discussion. Hume is rather arguing in general that convention gives rise to kinds of behaviors and it is governed by a “sense of general interest” and that such conventions are previous to the act of promising. He mentions by the end of the passage, that even human languages are established in the same manner, which is an important remark for our present topic since it was one of Coady’s claims concerning Hume’s reductionism. According to Coady, our trust in testimony depends, in Hume’s account, on particular individual observations. This would involve that there would be no prima facie or a priori reason for trusting testimony as such, and that, in turn, would imply the impossibility of learning a language: “The ability to make true reports with words is connected with using the words correctly and this ability is something that can only be exhibited in the consistent making of true reports.” (COADY, 1973 p. 154). It seems to me, however, to be systematic reasons to reject Coady’s criticism[11] in favor of Hume: a conventional way of practice that is based on a sense of common interest “gives us confidence of the future regularity of their conduct” (vid. supra).
(2) While there are prima facie reasons to trust human testimony based on convention, there is still an open question, namely, how and when particular testimony is to be epistemically evaluated as a reliable source of belief. I will sustain that the role of virtue in this task is based on two features of Hume philosophy:
(i) his rejection that rational and emotional aspects of human nature are independent of one another.
(ii) closely related to (i) his broad view of virtue, according to which under personal merit,
we shall consider every attribute of the mind, which renders a man an object either of esteem and affection, or of hatred and contempt; every habit or sentiment or faculty which, if ascribed to any person, implies either praise or blame, and may enter into any panegyric or satire of his character and manners. (EPM 1.1.9 SBN 174).
Hume’s definition comprises what would traditionally be called moral virtues, as well as intellectual abilities. The passage provides additional information that complements the thesis that that testimonial exchange is a conventional way of practice in Hume’s sense. First, there are conditions for the existence of testimony as such, which are independent of the epistemic appraisal of its reliability[12]. Assessing testimony is not simply assessing one´s beliefs, but also assessing the speaker’s “habit or sentiment or faculty which […] implies either praise or blame”. For, testimony as such to be a prima facie reliable source of belief is to say nothing further than human beings have an inclination to believe others reports because of convention. The fact that we have “an inclination to truth and a principle of probity” (vid. supra), because of a sense of common interest, are the underlying condition for testimony as such.
Second, granted that testimony can stand on its own feet, it is a rather different question how we come to a criterion for epistemic evaluation of particular cases of testimony. Reflective evaluation of testimony, that is, establishing the particular trustworthiness of the speaker is a matter of virtue in the broader humean sense. In this sense, however, relying on testimony can be considered dependent on empirical evidence.
My claim is that in Of miracles Hume is mostly concerned with the second question, while he takes for granted the first. Consider this passage:
And as the evidence, derived from witnesses and human testimony, is founded on past experience, so it varies with the experience, and is regarded either as a proof or a probability, according as the conjunction between any particular kind of report and any kind of object has been found to be constant or variable. There are a number of circumstances to be taken into consideration in all judgments of this kind; and the ultimate standard, by which we determine all disputes, that may arise concerning them, is always derived from experience and observation. Where this experience is not entirely uniform on any side, it is attended with an unavoidable contrariety in our judgments, and with the same opposition and mutual destruction of argument as in every other kind of evidence. […] This contrariety of evidence, in the present case, may be derived from several different causes; from the opposition of contrary testimony; from the character or number of the witnesses; from the manner of their delivering their testimony; or from the union of all these circumstances. We entertain a suspicion concerning any matter of fact, when the witnesses contradict each other; when they are but few, or of a doubtful character; when they have an interest in what they affirm; when they deliver their testimony with hesitation, or on the contrary, with too violent asseverations. (EHU 10.1.7).
Hume delivers a sort of catalog of things to take into consideration for reflective evaluation of testimony. It strikes first a required sensitivity to previous experience of the matter (if it constitutes a proof – v.g. all humans are mortal –, or a probability, – v.g. gray clouds bring rain). Hume affirms that evidence from testimony varies with evidence from experience. A second interesting remark is that when there is a mismatch between testimony and experience, the latter must be “the ultimate standard, by which we determine all disputes” (vid. supra). In the third place, Hume offers a list of several causes of “contrariety of evidence” (vid. supra), which include, the character and motivations of witnesses, presence of contradiction, etc. It seems to me fare to infer from the passage, that in the absence of causes to doubt people’s report, they are mostly credible. Hume´s epistemic concern is a question of rationality. As he himself points out, being a good reasoner, can still lead to false belief: “The Indian prince who refuses to believe the first relations concerning the effects of frost, reasoned justly” (EHU 10.1.9 SBN 113) because of his lack of experience of cold climates.
In conclusion, testimony as such is prima facie reliable. Testimonial belief in miracles requires special attention since it dramatically opposes experience. Reflecting upon the reliability of testimony in cases when there is opposing evidence makes epistemic agents reasonable, since it opposes two vices, namely, credulity, on the one extreme, and skepticism, on the other.[13] Traits of the witness’ character are also important, remember that “a man delirious, or noted for falsehood or villany, has no manner of authority with us” (vid. supra).
2.2 Justification of belief and epistemic virtue[14]
The first thing I want to emphasize about the notion of "virtue" is its "hinge" character between theoretical and practical philosophy in Hume’s work. Its relevance lies in the fact that the humean approach does not distinguish between two types of virtue as in the Aristotelian tradition, rather than, intellectual virtue is integrated with ethical virtue under the same description (THN 3.3.4.5-8). For Hume the "natural abilities" among which the various epistemic and doxastic virtues are founded (See THN 3.3.4: rapid apprehension, penetration, genius, true judgment, wisdom, good sense, etc.) as well as the moral virtues, respond in last term to what can be called pre-epistemic and pre-morals standards (utility and pleasure). This idea, systematically considered, is not far from his account of natural convention. Virtue is then determined by what is agreeable or useful to oneself or to others (EPM 9, THN 3.3.1.30). Hume affirms that "the approbation of moral qualities most certainly is not derived from reason or any comparison of ideas; but proceeds from a moral taste, and from certain sentiments of pleasure or disgust which arise upon the contemplation and view of particular qualities or characters." (THN 3.3.1.15 SBN 581[15]). However, as feelings are so fluctuating, he adds “in order, therefore, to prevent those continual contradictions, and arrive at a more stable judgment of things, we fix on some steady and general points of view; and always, in our thoughts, place ourselves in them, whatever may be our situation.” (THN 3.3.1.15 SBN 581-582). Such fiction is only possible through reflection whether it be aesthetic, moral or epistemic. The central point is that there is a fundamental difference between perceiving (or feeling) and judging virtue. Judging requires correction and reflection. Hume offers a quite prosaic example of this: “Our servant, if diligent and faithful, may excite stronger sentiments of love and kindness than Marcus Brutus, as represented in history; but we say not upon that account, that the former character is more laudable than the latter.” (THN 3.3.1.16 SBN 582). Rationality requires that "we seek some other standard of merit and demerit, which may not admit of so great variation" (THN 3.3.1.18 SBN 583).
It often appears in passages where we find these kinds of reflections, a
characterization of the "wise" and the "vulgar", where the
latter seems to be the one incapable of reflective assessment of her own
beliefs and feelings. A wise man not only “proportions his belief to the
evidence”, but he does it by a “refinement of reason”[16], which makes
her a virtuous reasoner. Epistemic evaluation is for Hume always a matter of
virtue, either concerning testimony, or any other kind of belief formation
mechanism. Notice that from the perspective of the wise “we frequently hesitate
concerning the reports of others” (EHU 10.1.6 SBN 112), while, in contrast,
from the perspective of the vulgar, “no weakness of human nature is more
universal and conspicuous than what we commonly call credulity, or a too easy
faith in the testimony of others” (T 1.3.9.12 SBN 112).
Credulity is a doxastic vice which is caused by a weakness of human nature, an error which is of the same kind with prejudice, as Hume’s famous example that “an Irishman cannot have wit, and a Frenchman cannot have solidity” (T 1.3.13.7). Experience of human nature, as well as observation of the general course of things[17] are “the true standard of this, as well as of all other judgments”, though “we seldom regulate ourselves entirely by it; but have a remarkable propensity to believe whatever is reported, even concerning apparitions, enchantments, and prodigies, however contrary to daily experience and observation” (T 1.3.9.12 SBN 113).
Despite this difficulty, Hume doesn't hesitate to affirm the relevance of testimonial evidence as a source of knowledge, since “there is no species of reasoning more common, more useful and even necessary to human life, than that which is derived from the testimony of men, and the report of eyewitnesses and spectators” (EHU 10.1.5 SBN 111).
This doesn't differ from his view concerning causal reasoning, which is, he holds, “not only a true species of reasoning, but the strongest of all others” (THN 1.7.7 - footnote SBN 96-97).
Final remarks
Of miracles is intended by Hume as a sound ground to the avoidance of superstition:
Nothing is so convenient as a decisive argument of this kind, which must at least silence the most arrogant bigotry and superstition, and free us from their impertinent solicitations. I flatter myself, that I have discovered an argument of a like nature, which, if just, will, with the wise and learned, be an everlasting check to all kinds of superstitious delusion, and consequently, will be useful as long as the world endures. (EHU 10.1.2, SBN 110).
This concern is also one of Hume's most important philosophical motivation both in his Treatise and Enquiry, his analysis of human cognition exposes both the lack of a rationalist final foundation for knowledge along with the danger of skepticism and, at the same time, the unavoidable character of our cognitive instincts. Among them, a natural propensity toward reasoning: curiosity, and an ambition “of contributing to the instruction of mankind” (T 1.4.7.12).
Since therefore it is almost impossible for the mind of man to rest, like those of beasts, in that narrow circle of objects, which are the subject of daily conversation and action, we ought only to deliberate concerning the choice of our guide, and ought to prefer that which is safest and most agreeable. And in this respect I make bold to recommend philosophy, and shall not scruple to give it the preference to superstition of every kind or denomination. For as superstition arises naturally and easily from the popular opinions of mankind, it seizes more strongly on the mind, and is often able to disturb us in the conduct of our lives and actions […] Generally speaking, the errors in religion are dangerous; those in philosophy only ridiculous. (T 1.4.7.13 SBN 271).
Similarly, a just reasoner ponders reliability of testimony “from the opposition of contrary testimony; from the character or number of witnesses; from the manner of delivering their testimony; or from the union of all these circumstances” (EHU 10.1.7 SBN 112). When it concerns the character of the witness, it occurs by “observing the variety of conduct in different men” (EHU 8.1.9 SBN 85). This evaluation a witness can take place only in terms of character traits that for Hume count as virtues. In Of miracles, as in many other texts, Hume's argument depends on the distinction between kinds of believers. This becomes evident from his permanent drawing the counterpoint between the wise and 'judicious' or 'learned' (EHU 10.2.2, SBN 120) who possesses “qualities of the mind” like “wit, good-sense, courage, integrity” (Abstract 659-660), and the vulgar or foolish. Those kinds of believers, I claim, embodied different belief formation mechanism that render their beliefs more reliable, this is why “we are not only better pleased with the approbation of a wise man than with that of a fool but receive an additional satisfaction from the former” (THN 2.1.11.12 SBN 321).
In conclusion, a belief has epistemic value as expression of a wise reasoner. Thus, for Hume, intellectual virtues as character traits dispose the reasoner to develop stable cognitive abilities or excellences (EPM App. 4 - none of which suffices on their own) and dispositions to form reliable beliefs, to speak the truth, and to assess other believers’ assertions.
Testimonio y virtudes intelectuales en la epistemología de Hume
Resumen: En este artículo, considero algunas cuestiones relacionadas con la epistemología del testimonio de David Hume. El eje central será la acusación de reductivismo e individualismo de los estudiosos en contra de la opinión de Hume sobre la evidencia testimonial. Dicha acusación se basa en su tratamiento de los reportes sobre hechos milagrosos, presente en la sección X de su Investigación sobre el entendimiento humano. En primer lugar, se abordan los argumentos en contra de la posición de Hume y algunas respuestas en la literatura reciente, para luego ofrecer una interpretación alternativa con respecto a la forma en que debería articularse esa defensa. Mi estrategia está estrechamente relacionada con la noción de virtud de Hume y el papel que desempeña en su epistemología, principalmente como se presenta en su Tratado de la naturaleza humana. Finalmente, abordaré el problema relacionado con la forma en que debe entenderse correctamente la sección “Of miracles” de la Investigación, ya que varios malentendidos sobre la epistemología del testimonio de Hume emergen parcialmente del carácter particular y el objetivo de dicho ensayo.
Palabras claves: David Hume. Testimonio. Milagros. Virtud. Epistemología.
References
ANSCOMBE, G. E. M. What is it to believe someone? In: DELANEY, C. F. (ed.). Rationality and religious belief. University of Notre Dame Press, 1979. p. 141-151.
COADY, C. A. J. Testimony and observation. American Philosophical Quarterly, v. 10, n. 2, p. 149-155, 1973.
COADY, C. A. J. Testimony: a philosophical study. Oxford: Clarendon Press, 1992.
FAULKNER, P. David Hume's reductionist epistemology of testimony. Pacific Philosophical Quarterly, v. 79, n. 4, p. 302-313, 1998.
GELFERT, A. Hume on testimony revisited. Logical Analysis and History of Philosophy, v. 13, p. 60-75, 2010.
GRECO, J. Agent reliabilism. Noûs, v. 33, n. 13, p. 273-296, 1999.
HARDWIG, J. Epistemic dependence. Journal of Philosophy, v. 82, n. 7, p. 335-349, 1985.
HUME, D. Essays on moral political and literary: Part One 1742. Part Two 1752. Edited by Eugen F. Miller. Indianapolis: Liberty Fund, 1742-1752.
HUME, D. A treatise of human nature (THN), 1739-1740. Edited by D. F. Norton and M. J . Norton. Oxford: Oxford University Press, 2000; SELBY-BIGGE, L. A.; NIDDITCH, P. H. Oxford at The Clarendon Press: London, 1978.
HUME, D. An enquiry concerning human understanding (EHU), 1748. Edited by Tom L. Beauchamp. Oxford University Press, 1999. Enquiries concerning human understanding (EHU) and concerning the principles of morals (EPM). Edited by L. A. Selby-Bigge. Third edition revised by P. H. Nidditch. Oxford, Clarendon Press, 1975, reprinted 2003.
O’BRIAN, D. Hume and the intellectual virtues. Discipline Filosofiche, v. 22, n. 2, p. 153-172, 2012.
POMPA, L. Human nature and historical knowledge: Hume, Hegel and Vico. Cambridge: Cambridge University Press, 1990.
ROOT, M. Hume on the virtues of testimony. American Philosophical Quarterly, v. 38, n. 1, p. 19-35, 2001.
SCHAFFER, K. Curious virtues in Hume's epistemology. Philosophers' Imprint, v. 14, p. 1-20, 2014.
SOSA, E. A virtue epistemology: apt belief and reflective knowledge. Oxford: Oxford University Press, 2007. V. 1.
TRAIGER, S. Beyond our senses: recasting book I, Part III of Hume's treatise. Hume Studies, v. 20, n. 2, p. 241-259, 1994.
TRAIGER, S. Doxastic virtues in Hume’s epistemology. Hume Studies, v. 35, n. 1-2, p. 211-229, 2010.
VITZ, R. Doxastic virtues in Hume’s epistemology. Hume Studies, v. 3, n. 1-2, p. 211-229, 2009.
WELBOURNE, M. Is Hume really a reductivist? Studies in History and Philosophy of Science, v. 33, p. 407-423, 2002.
WILLIAMS, B. Deciding to believe. In: Problems of the self: philosophical papers 1956-1972. Cambridge: Cambridge University Press, 1973.
ZAGZEBSKI, L. Virtues of the mind: an inquiry into the nature of virtue and the ethical foundations of knowledge. Cambridge: Cambridge University Press. 1996.
[1] Assistant Professor, Departamento de Humanidades, Facultad de Educación, Universidad Andres Bello, Santiago – Chile. https://orcid.org/0000-0001-9941-1974 E-mail: ruth.espinosa@unab.cl.
I would like to thank Dr. Luis Placencia, Prof. Rainer Enskat and Dr. James Mattingly for their helpful comments on a previous version of this article. I would also like to thank Dan O'Brian for providing me with a copy of his paper, which I cite below.
[2] Coady’s treatments of Hume’s account of testimonial based belief is to be found in his early paper of 1973, “Testimony and Observation”, and is reproduced with minor changes in his book of 1992, “Testimony: a philosophical study”.
[5] Although assertion is the way of expression of belief, it doesn’t follow from that that assertion is always expression of belief. See Williams (1973, p. 137).
[8] On this see Pompa (1990, especially p. 34ff and 192). The same idea can be found in Foaulkner (1998), he states that Hume´s account of testimony is rather naturalistic. Also Traiger (1994). Both conditions can be seen together in many passages of Hume’s Treatise, especially eloquent is THN, SBN 113.
[10] Hume enumerates many more conditions: “This contrariety of evidence, in the present case, may be derived from several different causes; from the opposition of contrary testimony; from the character or number of witnesses; from the manner of delivering their testimony; or from the union of all this circumstances.” (EHU 10.1.7).
[11] On this aspect of Hume’s account of testimony, I distance myself with Paul Foulkner’s claim that although Hume is actually a reductionism, his reductionism was misunderstood by Coady (FAULKNER, 1998).
[12] Saul Traiger has argued that Anscombre and Coady have committed an interpretative mistake by attributing “an account of justification of belief based on what Hume presents only as an account of the formation of belief” (TRAIGER, 2010, p. 46-47).
[13] Cf. O’Brian (2012) He note Hume’s use of the expression ‘due medium’, but unless Aristotle, in Hume such is determined by utility, due to his naturalistic account of virtue.
[14] A catalog and detailed treatment of doxastic, intellectual and epistemic virtues in Hume can be found in O’Brian (2013) and Schaffer (2014). The interest in the doctrine of the intellectual virtue of Hume has been awakened to a great extent because of the relevance that the epistemology of virtue has acquired in the contemporary epistemological panorama (SOSA, 2007; GRECO, 1999; 2010; ZAGZEBSKI, 1996). In this way, the humean vision of intellectual virtue has been pointed out by some authors (vid. Vitz 2009) as a possible contribution to the debate that prevails in the epistemology of contemporary virtue and whose main lines are the so-called "virtue reliabilism" (v.g. Sosa) of an externalist character and the "virtue responsibilism" (v.g. Zagzebski), a name that designates those theories modeled according to the Aristotelian vision of ethical virtue as a character trait.
[15] Al passages in this paragraph are quoted from section 1, whose title is “of the origin of the natural virtue and vices”.