Thomas Reid, the Internalist

1.1 Identifying justification in Reid

There are a number of challenges to locating a view of epistemic justification in Reid. To begin with, Reid never provides a sustained discussion of the concept of epistemic justification. In all of his works, he uses ‘justified’ only four times and ‘justification’ merely once [23, 77, 89, 358, 398].⁴ Only three of these occurrences obviously enlist the epistemic sense of justification, and all of these focus on particular cases of justified belief. Here is one such instance:

A child who has once burnt his finger, by putting it in the flame of one candle, expects the same event if he puts it in the flame of another candle, or in any flame, and is thereby led to think that the quality of burning belongs to all flame. This instinctive induction is not justified by the rules of logic, and it sometimes leads men into harmless mistakes, which experience may afterwards correct; but it preserves us from destruction in innumerable dangers to which we are exposed. [398]

Little can be inferred from this passage. Reid makes it clear that induction is not justified by logic, but does not address whether such inductive beliefs are ultimately justified. Reid’s other mentions of justified beliefs are equally sparse, providing meager resources to flesh out a theory of justification:

If we should grant to Mr. Hume that our ideas of memory afford no just ground to believe the past existence of things which we remember, it may still be asked how it comes to pass that perception and memory are accompanied with belief, while bare imagination is not. Though this belief cannot be justified upon his system, it ought to be accounted for as a phenomenon of human nature. [352]

Here, Reid discusses what can be justified on the basis of Hume’s account of memory, but this provides nothing that will help in adjudicating the dispute between internalists and externalists. In order to establish what Reid’s view of the contemporary debate on justification might have been, we will have to look elsewhere.

One strategy for assessing what Reid might have thought about contemporary discussions of justification is by starting with the theoretical roles that epistemic justification is meant to play and then identifying which concepts in Reid are capable of filling those roles. The current notion of justification is meant to satisfy a number of theoretical roles, including making sense of truth as the aim of belief, capturing when rational agents are praiseworthy and blameworthy for their beliefs, and providing guidance on what beliefs to hold. The challenge, not just in Reid but also in the current debate, is that not all theorists agree about the relevance of these roles for a theory of justification. Process reliabilism prioritizes the thought that justification should be connected to truth, but there are others that argue that no sense can be made of justification being truth-conducive.⁵ Epistemic deontologists take epistemic blamelessness to be coextensive with justification, but many epistemologists hold the two notions can come apart.⁶ Proponents of naturalized epistemology think that the analysis of justification should offer guidance for epistemic improvement, while other epistemologists see this as inessential for theorizing about justification.⁷

Because all of these theoretical roles are up for debate, the role that I will be using to pinpoint justification in Reid is even more fundamental. Both internalists and externalists about epistemic justification agree that there are normative statuses that believers might satisfy or fail to satisfy, and that being justified in one’s beliefs is one such status. These statuses can be identified via discussions about what rational agents epistemically ought to believe. Whatever the views of internalists and externalists on epistemic praise and blame or justification’s truth-connection, they can at least agree that there is a sense of epistemic ought or should that attaches to the concept of justification. My project, then, will be to focus on this theoretical role, examining whether the normative epistemic statuses found in Reid are more amenable to an internalist or externalist treatment. From the outset, any epistemic sense of ought or should that attaches to belief is a candidate for Reid’s views on justification.

To see what this methodology might look like in practice, let’s consider two properties that beliefs might possess, being instinctual and having a just ground. Based on Reid’s comments, instinctual belief does not seem to be a normative epistemic status. Consider, for instance, the following passage:

When we consider man as a rational creature, it may seem right that he should have no belief but what is grounded upon evidence, probable or demonstrative; and it is, I think, commonly taken for granted, that it is always evidence, real or apparent, that determines our belief. If this be so, the consequence is, that, in no case, can there be any belief, till we find evidence, or, at least, what to our judgment appears to be evidence. I suspect it is not so; but that, on the contrary, before we grow up to the full use of our rational faculties, we do believe, and must believe, many things without any evidence at all […] We are irrational animals for a considerable time before we can properly be called rational […] If there be any instinctive belief in man, it is probably of the same kind with that which we ascribe to brutes, and may be specifically different from that rational belief which is grounded on evidence. [548]

Here, Reid contrasts the sorts of beliefs that we must believe, simply from a prudential point of view, and those beliefs that can be evaluated using a normative epistemic status. Reid points out that instinctive beliefs are formed before we are rational creatures, with many of them based on no evidence whatsoever. These beliefs are not ones that Reid thinks we should evaluate from the epistemic point of view, as they are ‘specifically different from that rational belief which is grounded on evidence.’ If there is any normativity that attaches to instinctual beliefs, those we believe while we are still irrational, it is purely a practical or prudential kind of normativity. Reid thinks that ‘it is necessary for our preservation, that we should believe many things before we can reason’ [333], but if instinctual beliefs are adopted merely out of practical necessity, this is not a place to go looking for Reid’s views on epistemic justification.

While the previous passage on instinctual beliefs focused on things we believe before we are rational creatures, it also mentions beliefs that are rationally formed, saying that, ‘when we consider man as a rational creature, it may seem right that he should have no belief but what is grounded upon evidence.’ Here we find our first normative epistemic status. A property that looks like a better candidate for a theory of justification in Reid is having good evidence, also known as having a just ground:

I shall take it for granted that the evidence of sense, when the proper circumstances concur, is good evidence, and a just ground of belief […] All good evidence is commonly called reasonable evidence, and very justly, because it ought to govern our belief as reasonable creatures. [328]

Here, we have normative language associated with a distinctly epistemic domain of evaluation. Because we are reasonable creatures, we ought to govern our beliefs based on when we have a just ground for belief, making having a just ground a candidate for a theory of justification. This is, of course, just a methodological preview, and we will discuss further what having a just ground involves in Section 3, but this has served as a helpful primer on identifying justification in Reid. In this paper, I will argue that there are two normative epistemic statuses that can be found in Reid, being blameless and having a just ground. If Reid is an externalist concerning epistemic justification, then at least one normative status for belief will be analyzable in a manner friendly to externalists. I will argue, however, that both statuses are best characterized as accessibility accounts of justification.

1.2 Access internalism

Even though Reid is often interpreted as an externalist about justification,⁸ he actually subscribes to access internalism, the clearest form of internalism. It is a fraught question within epistemology precisely what distinguishes an internalist from an externalist theory of justification, and I will not try to provide necessary and sufficient conditions to adjudicate the issue.⁹ Instead, I will focus on two points of agreement between internalists and externalists. We have already seen one of these points. Both internalists and externalists agree that there are normative statuses that believers might satisfy or fail to satisfy, and that being justified in one’s beliefs is one such status. These statuses can be identified by the presence of normatively loaded language surrounding believers and their mental lives, along with discussions about what rational agents should or ought to believe. Whatever the views of internalists and externalists on the necessary and sufficient conditions for justification, they can thus at least agree that there is a sense of epistemic ought or should that attaches to the concept of justification. My project, then, will be to focus on this feature of justification, examining whether the normative epistemic statuses found in Reid are more amenable to an internalist or externalist treatment. If Reid is an externalist concerning epistemic justification, then at least one of being blameless or having a just ground will be analyzable in a manner friendly to externalists.

The second point of agreement between internalists and externalists that I will leverage is that accessibility accounts of justification are clearly internalist. Wherever the line is drawn between internalism and externalism, access internalism will always fall on the internalist side of that line. Let us take a fairly simple account:

Access Internalism: All of the facts that are required to justify an agent S in believing p are internally accessible to S

What precisely are these facts? What does it mean for them to be accessible? When I reflect, I can discover many things about my mental life. I can take stock of my memories and current experiences, and I can ascertain the strength with which certain things seem to be the case. These are the sorts of facts that can justify my beliefs—that I remember that p, I am experiencing that p, or that p seems to be the case. The central thought is that these are facts that I can tell whether or not they obtain. Now there might be various ways in which I reflect on these facts about my mental life. I could have examined them in the past, I could currently be reflecting on them, or I might never have thought twice about them. Which choice we make here will yield different forms of access internalism. I will choose the latter for now and, in Section 3.3, we will see reason to think that this is a fitting choice. Thus, the sense in which the facts that justify S in believing p must be accessible is that they must be accessible upon reflection. These facts need not have ever been accessed before in order for them to do their justifying work—the agent merely must be capable of reflecting on them.

3.1 Evidence and the distorting influence of prejudice

For Reid, it is desirable not only that our beliefs be blameless, but that they also have a just ground. The following passage identifies another normative status beliefs might satisfy:

I shall take it for granted that the evidence of sense, when the proper circumstances concur, is good evidence, and a just ground of belief […] All good evidence is commonly called reasonable evidence, and very justly, because it ought to govern our belief as reasonable creatures. [328]

Here we get a window into how Reid thinks that the evidence we have should regulate our beliefs. Good evidence provides us with just grounds that ‘ought to govern our belief.’ This good evidence, however, can only provide one with a just ground when the proper circumstances occur. What are these proper circumstances? There are a couple of possibilities. On the one hand, it could be that the appropriate circumstances that Reid is indicating are the previously mentioned misinterpretations of sense experience. The appropriate circumstances would then be when we do not infer propositions beyond what our experiences support. If this is the case, though, then just grounds will be of no help to the externalist, for having a just ground will come to the same thing as believing blamelessly. On the other hand, and this is the interpretation that I favor, Reid might have in mind his other thoughts on evidence, discussions to which we now turn.

We are all capable of incorrectly judging the strength of various pieces of evidence. How, then, can we know what the actual strength of the evidence is? Reid’s answer is that the strength of the evidence can be identified with how strong the evidence would seem to particularly clear-thinking individuals. Reid uses this criterion to determine both the strength of evidence:

I think, in most cases, we measure the degrees of evidence by the effect they have upon a sound understanding when comprehended clearly and without prejudice. [482]

And when something is self-evident:

Self-evident propositions are those which appear evident to every man of sound understanding who apprehends the meaning of them distinctly, and attends to them without prejudice. [282]

We can see how strong evidence is by how it affects someone who is free from prejudice.¹⁴ These are the ‘proper circumstances’ Reid has in mind when he discusses having a just ground for one’s belief. Immediately before saying that the evidence of sense could constitute a just ground for belief, Reid makes reference again to the absence of prejudice:

What […] evidence is, is more easily felt than described […] every man of understanding can judge of it, and commonly judges right, when the evidence is fairly laid before him, and his mind is free from prejudice. [328]

What makes the evidence of sense a just ground for belief is exactly what allows one to accurately assess the strength of evidence—a sound understanding and freedom from prejudice. This directive goes beyond Reid’s discussion of blamelessness in regulating one’s beliefs and claims something further, that there is a certain kind of evidence that ought to regulate our beliefs.

One worry at this point might be that evidence in Reid is not a normative notion at all, calling into question whether having a just ground is actually a candidate for an account of justification in Reid. William Alston, for example, has pointed out that there are places in Reid where evidence seems to be a merely psychological notion. Consider, for example, the following description of what unites the different kinds of evidence:

I confess that, although I have, as I think, a distinct notion of the different kinds of evidence above mentioned […] yet I am not able to find any common nature to which they may all be reduced. They seem to me to agree only in this, that they are all fitted by nature to produce belief in the human mind. [328]

As Alston points out, what unites different kinds of evidence in this passage is just ‘that we are so constituted that they produce beliefs in us’ (1985: 438). If this is right, then we can identify evidence just by its effect of prompting us to believe, providing a purely psychological characterization of evidence.

Although there are passages in Reid where evidence appears to pick out a psychological notion, there is also ample evidence that Reid also takes evidence to be normative. In numerous places, Reid evaluates evidence as good [328, 345, 379, 517, 541, 626] or sufficient [328, 345, 379, 517, 541, 626], and says that evidence ‘ought to govern our belief as reasonable creatures’ [328]. Reid also characterizes believing without evidence as a ‘weakness which every man is concerned to avoid,’ [328] an impossibility if evidence just is what compels belief.¹⁵ Perhaps the best answer, then, for whether Reid’s notion of evidence is psychological or normative is that it is both. Evidence often compels assent in us, and when that evidence is good, it also supplies a just ground. This dual view is eventually what Alston settles on as well, ultimately agreeing that Reid’s account of evidence has both psychological and evaluative components.¹⁶

3.2 Foundationalism and accessibility

Given that being free from prejudice is a crucial element of having a just ground, an important question is whether being free from prejudice is reflectively accessible. Given what else Reid says about prejudice, this certainly seems to be the case. Reid gives epistemic priority to the beliefs that are produced naturally by our constitutions, but these original judgments can be obscured by other beliefs that we have acquired along the way:

It must therefore require great caution, and great application of mind, for a man that is grown up in all the prejudices of education, fashion, and philosophy, to unravel his notions and opinions, till he find out the simple and original principles of his constitution, of which no account can be given but the will of our Maker. [99]

Here we see a few sources of prejudice—education, fashion, and philosophy—and Reid describes further the specific prejudices that come with education in the special sciences and philosophy [20, 282–83]. Being under the influence of such prejudices, though, is not without remedy. It takes great caution and attention, but a man can rid himself of prejudice so that he can discern the beliefs he has due to his original constitution. How does one go about ridding oneself of prejudice? By reflecting on what beliefs are actually produced by the natural faculties, and as we have seen before, these beliefs are marked by their irresistibility. The ability to rid oneself of prejudice is derivative of a more generality ability, the ability to directly apprehend what is self-evident:

The same degree of understanding which makes a man capable of acting with common prudence in the conduct of life, makes him capable of discovering what is true and what is false in matters that are self-evident, and which he distinctly apprehends. [522]

Because of this ability, rational agents are capable of ridding themselves of prejudice to uncover what is self-evident. This is not to say that the process is not a difficult one, as Reid admits that ‘it is not impossible that what is only a vulgar prejudice may be mistaken for a first principle. Nor is it impossible that what is really a first principle may, by the enchantment of words, have such a mist thrown about it, as to hide its evidence, and to make a man of candor doubt it’ [231]. There are moments where we may mistake a mere prejudice for a first principle, or where a first principle may be so obscured as to incur doubt. Nevertheless, despite these difficulties, the kind of understanding needed to act with common prudence makes it possible to root out our various prejudices.

All of this fits effortlessly within Reid’s foundationalism. Self-evident beliefs form the foundation of a subject’s justification, and then further justified beliefs can be derived from this foundation, shedding light on deeply held prejudices along the way. This is what Reid understands himself to have done concerning the Theory of Ideas. Reid says that, at one point, he ‘believed the whole of Berkeley’s system,’ something he was only freed from by ‘candidly and impartially […] seeking for the evidence of this principle’ [283]. What Reid found was that beliefs about the material world are irresistible, thus firmly establishing them as a just ground of belief:¹⁷

This conviction is not only irresistible, but it is immediate, that is, it is not by a train of reasoning and argumentation that we come to be convinced of the existence of what we perceive. [259]

As Plantinga points out, Reid is a nonclassical foundationalist in that Reid regards foundational beliefs to extend to propositions about the external world.¹⁸ It is self-evident that there is a material world, and it follows from this that the Theory of Ideas is false, revealing Reid’s original confidence in the Theory to be mere prejudice. On Reid’s account then, freeing ourselves of prejudice is indeed reflectively accessible. By reflecting on what beliefs we hold irresistibly, we can rid ourselves of prejudice and have a just ground for our beliefs.

3.3 Irresistible versus involuntary

One issue that is important to address is the difference between beliefs that are irresistible versus ones that are merely involuntary. We have already discussed that Reid thinks that all beliefs are involuntary, but he takes irresistibility to be an even stronger notion, with a number of dissimilarities between involuntary and irresistible beliefs. Beliefs are involuntary in that they arise spontaneously upon examining evidence, but irresistible beliefs possess a stronger involuntariness—they are believed no matter what evidence one examines. Such beliefs are ‘principles which irresistibly govern the belief and the conduct of all mankind’ [110, italics mine]. But what about the philosophers who have subscribed to the Theory of Ideas? Haven’t they managed to throw off their commensensical beliefs? Reid holds that, even for them, rejecting these beliefs is impossible:

Leaving [the Theory of Ideas], therefore, to those who have occasion for it, and can use it discreetly as a chamber exercise, we may still inquire how the rest of mankind, and even the adepts themselves, except in some solitary moments, have got so strong and irresistible a belief, that thought must have a subject, and be the act of some thinking being. [110]

Such are irresistible beliefs, that they cannot be given up by anyone regardless of their evidence. Prejudices, meanwhile, are involuntary in that they are beliefs, but they are not held come what may. Prejudices have been learned, and thus they can also be given up. The way Reid himself came to give up his philosophical prejudices was by realizing their conflict with self-evident first principles:

I once believed this doctrine of ideas so firmly as to embrace the whole of Berkeley’s system in consequence of it; till, finding other consequences to follow from it, which gave me more uneasiness than the want of a material world, it came into my mind, more than forty years ago, to put the question, what evidence have I for this doctrine that all the objects of my knowledge are ideas in my own mind? From that time to the present I have been candidly and impartially, as I think, seeking for the evidence of this principle, but can find none, excepting the authority of philosophers. [283]

One does not have direct control over one’s beliefs, but by examining the evidence, one can find one no longer holds the prejudices one once did. For Reid, this occurred by examining where the consequences of the Theory of Ideas and self-evident propositions came into conflict with one another.¹⁹ Thus, it is compatible to hold that rational agents can rid themselves of prejudice as well as that these prejudices are believed involuntarily.

Let’s do some bookkeeping. We have identified a second normative status in Reid which beliefs might satisfy, that of having a just ground. One’s belief that p has a just ground when it is based on good evidence, evidence that would be considered adequate by a person free from prejudice. Given that it is possible for a person to rid themselves of prejudice by reflection, this is an access internalist condition for having a just ground for one’s belief. What justifies one’s beliefs begins with what is self-evident, and access to self-evident propositions can be attained by clearing away the distorting effects of prejudice. It is important to note that it is not freedom from prejudice itself that justifies an agent’s beliefs—that task is left to the believer’s evidence. What the lack of prejudice does is allow a subject to see their evidence aright, making whether their evidence supports a particular proposition accessible by reflection.

Reid’s description of how one becomes free of prejudice places some limitations on which version of access internalism he can endorse. Clearly, for Reid, being free of prejudice does not imply that one has actively gone through one’s stock of evidence and examined all of one’s prejudices. After all, ‘a man who never speculated about evidence in the abstract may have a good judgment’ [328]. Furthermore, having a general ability to rid oneself of prejudice does not ensure that this ability can be exercised at any moment whatsoever. For there may be times where the strength of our prejudice blinds us to our own biases—such is the pernicious nature of prejudice: ‘But as light may be so offensive that the bodily eye is shut involuntarily, may not something similar happen to the eye of the understanding, when brought to a light too offensive to some favourite prejudice or passion, to be endured?’ [74]. So, if we are going to pursue an accessibility account of what provides a belief with just grounds, it will have to be an account where the relevant facts are just reflectively accessible in principle. This accords with our precisification of access internalism in Section 1.2. For Reid, having a just ground is accessible in precisely the access internalist sense. Our beliefs have a just ground when they appear evident to us when we are free from prejudice, and eliminating prejudice is within a rational agent’s reflective power.

3.4 Having a sound understanding

One concern about this internalist interpretation of having a just ground might be that having a just ground not only requires that someone evaluate evidence without prejudice but also that they have a sound understanding: ‘In most cases, we measure the degrees of evidence by the effect they have upon a sound understanding when comprehended clearly and without prejudice’ [482, italics mine]. If having a sound understanding is a necessary condition for having a just ground for belief, but it is not possible to tell by reflection when one has a sound understanding, then perhaps Reid is best interpreted as an externalist after all. In order to evaluate this objection, let’s first get a better handle on what Reid might have in mind as a ‘sound understanding.’ To begin with, a sound understanding is also a freedom from certain mental defects. The first examples that come to the minds of contemporary readers might be cases of mental illness, but Reid brings more commonplace concerns to the fore. Amongst the disorders of the understanding, Reid lists inordinate trust in authority [489], excessive analogical reasoning [470], and emotional irrationality [473]. These defects are correctable in that one can be made aware of the disorder as well as its remedy:

When we know a disorder of the body, we are often at a loss to find the proper remedy; but in most cases the disorders of the understanding point out their remedies so plainly that he who knows the one must know the other. [488]

Thus, just like prejudice, not only can one be made aware of one’s disorders, but one can also rid oneself of them. Reid even goes so far as to identify these disorders as prejudices, calling them ‘biases of the understanding,’ [469] and ‘prejudices which have their origin from […] the constitution of human nature […] [and also] which arise from the particular way in which a man has been trained.’ [473] So a sound understanding is in part being devoid of intellectual defects.

It might be worried that, if Reid thinks that a sound understanding is simply being devoid of prejudice, then his evidence formulas can only be interpreted as redundant, making mention of freedom from prejudice twice over. This is an interpretation that we should only favor if there is no other way to construe Reid’s uses of ‘sound understanding.’ Fortunately, other explications are available. There are a couple things to point out here. On the one hand, Reid associates having a sound understanding with not having any intellectual defects. Being free from defects of this sort, and having properly functioning cognitive capacities, is not always reducible to forms of prejudice. Take, for example, Derek Brookes’s interpretation of sound understanding in Reid. On Brookes’s view, ‘by the condition of “sound understanding,” Reid means, in part, that the powers by which the belief is produced must be functioning in such a way as to bring about at least one particular end for which they were designed, namely, the production of true beliefs’ (1996: 8). If this is right, then having a sound understanding would go beyond the mere absence of prejudice, requiring that a subject has properly functioning belief-forming mechanisms.

Another thing worth noting is that Reid often discusses having a sound understanding in connection with the maturity of the intellect. Reid speaks often of the ‘ripeness of the understanding’ [240, 362, 397, 419] even going so far as to treat the ripeness of the understanding as equivalent to a sound understanding:

But the power of judging self-evident propositions, which are clearly understood, may be compared to the power of swallowing our food. It is purely natural, and therefore common to the learned and the unlearned, to the trained and the untrained. It requires ripeness of understanding, and freedom from prejudice, but nothing else. [434]

Whereas before Reid said that a sound understanding and freedom from prejudice are required to discern self-evident propositions [282], here he uses the ripeness of the understanding alongside the absence of prejudice. Why might the maturity of the intellect be important for comprehending evidence? According to Reid, one can only apprehend evidence if one is capable of forming judgments about what is true and false, and this power only comes with a certain degree of development: ‘I restrict [the power of judgment] to persons come to the years of understanding, because it may be a question whether infants, in the first period of life, have any judgment or belief at all’ [414]. So having a sound understanding is not just being free from mental defect but having a maturity that allows one to judge whether a proposition is true or false, as this is a necessary condition of discerning evidence. Furthermore, having a sound understanding enables one to reflect on one’s own beliefs:

The power of reflection upon the operations of their own minds, does not appear at all in children. Men must be come to some ripeness of understanding before they are capable of it. [240]

Thus, maturity of the intellect is important for having a just ground for two reasons. Firstly, because making a judgment and apprehending evidence is only possible if one has attained the requisite intellectual development, and, secondly, because it enables believers to reflect on their own epistemic states.

Combining these two strains in Reid, we get the result that a sound understanding refers to mature, properly functioning cognitive capacities. So, is having a sound understanding an externalist condition on having a just ground? The first element of having a sound understanding, being free from mental defect, is clearly not, as this is just as amenable to reflective correction as are prejudices. Having well-formed belief forming mechanisms, on the other hand, is mentally inaccessible and outside one’s reflective control. Given that having a sound understanding is part of what it takes for a belief to have a just ground, then isn’t having a just ground an externalist condition of justification after all?

In order to show that Reid is an externalist about justification, we would need to show that there are aspects of Reid’s views that cannot be accommodated by internalists. When it comes to Reid’s view on a sound understanding, though, we can understand this in a way that fits into the internalist picture. On all forms of internalism, it must be the case that my brain developed normally in order for me to have justified beliefs, indeed, for me to have any beliefs at all. I did not have any control over whether this mental development obtained or not. This fact, however, does not vitiate internalism because my brain developing normally is not part of what justifies my belief that p. Instead, on the accessibilist view, it is that my belief is based on what is reflectively accessible that justifies it. Similarly, it is necessary that I develop a sound understanding in order to have just grounds for my belief. But just because I do not have control over whether I develop a sound understanding does not mean that I don’t have access to what provides me with a just grounds for believing that p. On Reid’s account, we do have access to what provides a just ground for our beliefs, evidence that is discerned without prejudice. Having the requisite mental development is a mere precondition for me to be able to engage in the justificatory game at all, but this does not push Reid’s view into the externalist camp.

4.1 Identifying justification in Reid

Commentators on Reid weave together markedly different concepts in discussing his theory of justification. Keith Lehrer finds talk of justification in what he calls Reid’s meta-principle, the first principle that our faculties are not fallacious (1989: 162). According to Lehrer, our belief in this principle grants justification to our sensory beliefs, for then we have a reason to think that the beliefs produced by our faculties are true. James Van Cleve disagrees, however, arguing that this principle plays no special role in explaining why we are justified in holding perceptual beliefs. Van Cleve points out that the supposed meta-principle occurs seventh on a list of similar principles and is merely a summary of previous ones:

Consciousness is reliable;

Memory is reliable;

Perception is reliable;

All our faculties are reliable. (2015:360)²⁰

So, will focus on Reid’s ‘meta-principle’ get us any closer to his account of epistemic justification? The jury is still out.²¹

Another strategy takes Reid to equate justification with fulfilling one’s epistemic obligations. Blake McAllister explores this route, arguing that what Reid has in mind by justification is not violating one’s intellectual duties:

The position I will defend is that Reid [holds] that we are rationally permitted to believe in first principles, which, on my interpretation, amounts to justification […] One is rationally permitted to believe something if believing it does not violate the normative standards or duties that ought to regulate our cognition as intellectual agents […] We are not in violation of our duties if we are not blameworthy or cannot be criticized for believing as we do. (2016: 319)²²

McAllister argues that, for Reid, one is justified in believing that p if one is not blameworthy for so believing. But Van Cleve disagrees that this is what is even at issue. In his commentary on Reid, Van Cleve concedes that certain beliefs are blameless for Reid but denies that this status is to be equated with justification:

Very well; we have no obligation to stop believing as we do; we are permitted to keep on believing those things we cannot help believing. But even if this permission is of a distinctively epistemic sort […] it seems to me to fall short of what epistemologists typically mean by warrant or justification. (2015: 335)

After acknowledging that there is a sense of epistemic obligation in Reid on which one could be blameless in one’s beliefs, Van Cleve denies that this amounts to Reid’s account of justification, claiming that justification comes apart from mere blamelessness in contemporary epistemology. This claim is not without merit—as we have already seen, several externalists make precisely the claim that justified beliefs are not coextensive with blameless beliefs.²³

So, not only do we have disagreement about whether Reid is an internalist or externalist, but we have disagreement about how to even get a conceptual grip on epistemic justification in Reid. Is a belief justified if it is appropriately connected to a meta-principle? If one has satisfied one’s epistemic obligations?²⁴ My account of Reid does not deal in such ambiguities. Justification is normative—believers ought to have justified beliefs and thus the normative statuses for belief are crucial to identifying justification in Reid. By focusing on such normative statuses, we can avoid confusion over how to identify justification in Reid.

4.2 An improved internalist interpretation

Mine is not the first interpretation on which Reid is an internalist—Keith Lehrer famously holds this position as well. Nothing that I have said is incompatible with Lehrer’s position, but one advantage of my view is that it does not grant any important role to the following disputed passage:

If any truth can be said to be prior to all others in the order of nature, this seems to have the best claim; because, in every instance of assent, whether upon intuitive, demonstrative, or probable evidence, the truth of our faculties is taken for granted, and is, as it were, one of the premises on which our assent is grounded. [447]

Lehrer takes this passage to indicate that Reid is not a reliabilist. This is because Reid seems to be implying that we cannot have any evidence from the outputs of our faculties without also using as one of our premises that our faculties are reliable (Lehrer 1989: 162–63).²⁵ Van Cleve, on the other hand, insists that Reid should not have said this because it contradicts his other views. If Reid really thinks that a person can be justified in their perceptual beliefs without inductively arguing from the reliability of their faculties, then this passage is a mistake.²⁶

Fortunately, my view does not need to take sides on whether Reid should have said this or not. With or without it, the view that an agent’s beliefs are justified because of something to which they have reflective access still has sufficient textual support. On my view, a belief’s being produced by a reliable faculty is neither necessary nor sufficient for its being blameless or having a just ground. Against the necessity of reliability, it is possible that some contingent beliefs produced by a faculty might be irresistible, making these beliefs blameless and providing them with a just ground, and yet also be false. This is possible on a local scale; Reid holds that even if our faculties are sometimes mistaken, this does not make them fallacious [485]. In this case, the beliefs produced by this faulty mechanism would have a just ground even though they were not formed by a perfectly reliable process.

This point, however, is not decisive against a reliability requirement, for reliabilists do not require that a belief-forming process be completely accurate in order to justify beliefs. So, what about the global case? Can our beliefs still have a just ground even if they are formed by processes which are less reliable than not? Reid would answer in the affirmative. If one has a sound understanding and examines the evidence without prejudice, then one has a just ground for their belief even if God has deceived us and ‘there is no remedy’ [30]. Against the sufficiency of reliability, it is also possible that a belief be produced by a reliable belief-forming process but the belief not have a just ground. Imagine a case where I have a belief that is reliably formed but the belief-formation process, even though reliable, is based on some prejudice. In such a case, I would not have a just ground for my reliably-formed belief. So, even though I do not disagree with Lehrer, my view can yield an anti-reliabilist verdict without appealing to a particular reading of the passage in question.

4.3 Explaining where externalist interpretations of Reid go wrong

Another advantage of the interpretation I have put forward is that it can bring further clarity to the debate over whether Reid is an externalist about justification. The first issue that externalist interpretations of Reid run into is failing to distinguish between two distinct notions of externalism in contemporary epistemology, externalist conditions for justification and externalist conditions for knowledge. Not disambiguating between the two has led to a fair amount of obscurity in debates over the sense in which Reid is an externalist. Some interpreters of Reid simply make it explicit that they are concerned with Reid’s understanding of knowledge instead of justification, commonly when discussing Reid’s reply to the skeptic:

John Greco –

According to Reid, … knowledge arises from the proper functioning of our natural, nonfallacious (i.e. reliable) cognitive faculties. (2004: 150)

James Van Cleve –

If Reid’s externalist epistemology is correct, we can at any rate know many of the things the skeptic says we cannot know—we can know things our knowing of which implies that the skeptic is wrong. (2008: 305)

Greco and Van Cleve here link Reid’s externalism to what we can know. This, however, does not settle the question about whether Reid is best interpreted as an internalist or externalist about justification, as a number of justification internalists have buoyed their accounts of knowledge against Gettier problems by incorporating externalist conditions into their analyses of knowledge.²⁷ Calling Reid an externalist in the context of discussing conditions on knowledge is thus somewhat misleading, as even internalists would qualify as externalists about knowledge. Reid may endorse an external condition on knowledge without thereby being a justification externalist.

To their credit, neither Greco nor Van Cleve claim that their interpretations of Reid settle the question of whether or not he is an externalist about justification. Other commentators on Reid, unfortunately, have not been as careful. Consider, for example, the reliabilist interpretation of Reid by William Alston. At the outset, Alston purports to be focusing on a reliability account of knowledge, saying, ‘I don’t want to suggest that Reid puts forward a reliability account of the nature of knowledge […] However, the fact that his epistemological first principles have to do exclusively with reliability strongly suggests that this is the sort of account he would give’ (1985: 437; italics are my own). Alston then proceeds, however, to focus on rational justification for the rest of the discussion, leaving it far from clear whether he thinks that Reid holds a reliability condition on justification or knowledge.

Another, more recent example is the reliabilist interpretation of Reid by Philip de Bary. After presenting some passages in which Reid discusses the healthy functioning of our perceptual faculties, de Bary declares, ‘this is unalloyed externalism; the instinctive beliefs of healthy people of all kinds tend towards truth, whether or not the believer be aware of this tendency’ (2002: 83). What does de Bary have in mind here by externalism? It is not obviously justification. On one hand, he says, ‘On pure externalism, these causes of error need only be operating in order for a believer not to know—they do not need to be known to be operating’ (2002: 85; italics are mine), but a page later says that ‘this adds up to an account on which common sense beliefs are prima facie justified’ (2002: 86; italics are mine). Again, it is unclear whether de Bary thinks that Reid is an externalist about knowledge, justification, or both. My understanding of Reid clarifies this ambiguity by explicitly outlining the notion of internalism and externalism at stake. I take no position on whether Reid endorses a fourth, externalist condition on knowledge but instead argue that Reid is a specific kind of accessibility internalist about epistemic justification.

The second mistake made by externalist interpretations of Reid is to take some element of his view in isolation and then argue that it implies that Reid is an externalist. Plantinga proffers the following perspective on Reid:

According to Reid, ‘we measure the degree of evidence by the effect it has on a sound understanding’ […] what he means, I think, is that what counts here is (at least in part) the effect of believing B with respect to believing A for someone who suffers from no cognitive defect, or deficiency, or dysfunction; someone whose (relevant) noetic faculties are functioning properly. (1993b: 164)

Plantinga takes it that, because evidence can only correctly be seen by one with a sound understanding, Reid is a proto-proper functionalist. As we have seen, though, having a sound understanding is what allows one to reflectively form beliefs, and this ability is crucial because of its role in enabling detection of prejudice. Plantinga doesn’t even consider this possibility, however, even omitting a crucial part of the quote. We cannot tell what the import of a particular piece of evidence is just because we have a sound understanding; rather, ‘we measure the degrees of evidence by the effect they have upon a sound understanding when comprehended clearly and without prejudice’ [482]. Thus, when it comes to giving a full interpretation of Reid’s thoughts on justification, Plantinga’s reading comes up short.

Michael Bergmann interprets Reid as suggesting a proper functionalism due to Reid’s view that the perceptual beliefs we form are contingent features of our natures. According to Reid, it is completely contingent that our sense of touch causes beliefs about the hardness of an object: ‘No man can give a reason why the sensations of smell, or taste, or sound, might not have indicated hardness’ [121]. Bergmann argues that this view of Reid’s is a short step away from externalism:

[This example] pushes us toward a proper function analysis of justification. For it suggests that the fittingness of a doxastic response depends, in some cases at least, on the species of the cognizer who has it. What is it about the species of a cognizer that determines such fittingness in those cases? The answer that immediately suggests itself is that what makes a belief a fitting unlearned doxastic response to an experience has to do with the way the belief-producing faculties of the cognizer in question are supposed to function. For clearly that is something that can vary from species to species. (2008: 59)

Bergmann claims that Reid’s account pushes us towards proper functionalism, but would Reid actually endorse proper functionalism? It is not clear the answer is yes. Reid simply observes that the way our senses provide us with evidence could have differed. Could the way that the senses give rise to perceptual beliefs differ in an individual within our species and still provide them with evidence? Reid is silent on this issue, but given the rest of Reid’s views, so long as that individual had a mature understanding and was able to reflectively free themselves from prejudice, we can infer that such an individual’s senses would have provided them with a just ground for their perceptual beliefs regardless of how the faculties of the rest of their species functioned.

Perhaps we should let Plantinga and Bergmann off the hook; after all, they only claim that their views are inspired or suggested by Reid’s.²⁸ There are other, more egregious, isolated readings, though, by those who claim to be expositing Reid’s own views. For de Bary’s externalist interpretation of Reid, he points to the fact that someone who has never speculated about evidence in the abstract can have a just ground for their belief. This allegedly makes Reid an externalist because reliable faculties can then produce justified beliefs without any reflection on those faculties by the believer:

There is no requirement from Reid, then, that believers be aware of the link between innateness and truth, still less that they reason out any link for themselves. First principles produce their ‘effect without ever being attended to.’ Believers come to know things ‘by their constitution’ provided only that the link in fact holds; and Reid assumes that it does in fact hold, in paradigmatically reliabilist fashion. (de Bary 2002: 84)²⁹

The problem with de Bary’s argument is that all of this can be true with Reid nevertheless being an internalist. As we have seen, even if Reid holds that some subjects are not aware of the principles of evidence, this does not entail that they do not have access to what makes their beliefs justified. Reid advocates that beliefs have a just ground in virtue of the reflective access we have to whether we are prejudiced, even though we do not need to make good on this access in every case. Reid can still be seen to be the access internalist that he is—justification does not require that someone has reflected on the evidence they have for their beliefs, but simply that they have the capability of doing so.

Another place where Reid has been misinterpreted is in his listing of the First Principles of Contingent Truths. The first of these principles affirms ‘the existence of everything of which I am conscious’ [578]. If taken in isolation, this principle seems to be stating that there is a certain connection between our mental states and the external world. This is how Alston takes the Principles, arguing that ‘the fact that [Reid’s] epistemological first principles have to do exclusively with reliability strongly suggests that this is the sort of account he would give’ (1985: 437). If Alston is right, then maybe what Reid is doing is affirming reliabilist principles of our faculties like Van Cleve (2008: 305) takes Reid to be doing. What Alston fails to recognize, however, is that the Principles are not just assertions about the correspondence between our perceptions and the external world but are themselves facts that are self-evident. When Reid lists the First Principles of Contingent Truths, he is not simply listing truths about our constitutions (even though he does take them to be truths). Rather, he is listing certain fundamental Principles that are self-evident and irresistible and thus can be believed blamelessly and with a just ground. Because Alston does not locate the Principles of Contingent Truths within this broader internalist framework, he ultimately concludes that they are evidence in favor of Reid being an externalist.

4.4 Between externalism and internalism?

Even though there are several externalist interpretations that fail to take Reid’s full corpus into account, Rene van Woudenberg’s “Between Externalism and Internalism” attempts to do justice to both the externalist and the internalist elements in Reid. Van Woudenberg begins by identifying positive epistemic statuses in Reid, and even though most of these statuses are internalist, he ultimately argues that at least some of these statuses are external conditions. This would make Reid a weak externalist about positive epistemic status, someone who thinks that at least one of the conditions required for positive epistemic status are external. If these external conditions are part of what makes a belief justified, then this would make Reid a weak externalist about justification as well.³⁰

What are the externalist conditions that van Woudenberg identifies? Van Woudenberg holds that there are three elements of Reid’s thought that at least come close to being external conditions: (1) That a belief is produced by a reliable natural faculty (2013: 78), (2) that a belief is supported by evidence of the senses (2013: 81), and (3) that a belief is supported by testimonial evidence (2013: 84).³¹ Of these three, van Woudenberg only regards (1) as being a thoroughgoing externalist condition. With (3), van Woudenberg thinks it is clear that, when a person remembers the testimony that supports their belief, (3) is an internal condition. He then entertains the possibility that, when a person has forgotten the explicit testimony but still recalls that there is some objective support for the belief, (3) might be a partially external condition. Ultimately, however, he concludes that ‘in both cases the condition is internal’ (2013: 84), leaving (1) and (2) as the only candidates for externalist conditions for positive epistemic status.

When it comes to (2), a belief being supported by evidence of the senses, van Woudenberg thinks that this condition is mostly internal. Van Woudenberg also considers Bergmann’s argument, that the contingent connection between particular sensations and the beliefs they produce put Reid in the externalist camp. In response, van Woudenberg distinguishes between two different questions, ‘(A) whether we can tell by reflection alone whether certain evidence of the senses is present; and (B) whether we can tell by reflection alone why it is that sensations of a certain kind signify certain qualities, and why it is that those sensations induce us to believe that the objects perceived posses those qualities,’ arguing that only (A) is needed for Reid’s view of the evidence of the senses to qualify as internalist (2013: 82). In considering this point, van Woudenberg says, ‘Reid’s point is that there is no necessary connection between sensations and what they signify—and no necessary connection between this sensation (sign) and the consequent belief that something has a certain quality. What does this mean for condition [(2)]? Does it render [(2)] an external condition after all? It does not […] It does not turn [(2)] into an external condition, as it requires only that the subject can answer (A)’ (2013: 82).³² Van Woudenberg then goes on to argue that, in most cases, Reid would answer (A) in the affirmative, making Reid an internalist about condition (2).

The only point at which van Woudenberg thinks that (2) might be an external condition is when it comes to the difficulty of separating between perception and inference, a challenge we discussed in connection with blamelessness in Section 2. Van Woudenberg develops this argument by pointing out that we rarely separate between the beliefs we form about the emotions of others and the expressions that we observe on their faces, arguing that ‘most often we are unaware of our visual sensations, even upon reflection’ (2013: 82). In response to this suggestion, it will once again be helpful to reiterate that the difficulty of separating between perception and inference does not necessarily undermine an internalist understanding of Reid. If it is impossible in principle to distinguish between the information of the senses and the further conclusions we reach using such evidence, then we cannot portray Reid as subscribing to access internalism, but this is not what we find in Reid. Instead, even though Reid acknowledges that ‘it is unnatural to the mind to stop at the visible figure, and attend to it’ [146], he also does not regard it as impossible to do so, as ‘an excellent painter or statuary can tell, not only what are the proportions of a good face, but what changes every passion makes in it’ [147]. Even though it may be challenging to separate between what we perceive and what we infer from what we perceive, it is not impossible to do so, preventing (2) from conclusively being an externalist condition of positive epistemic status.

This leaves us with condition (1), that a belief is produced by a reliable natural faculty, as the crux of van Woudenberg’s argument. There is no disputing that this is an external condition. We cannot discern via introspection whether our beliefs are formed by reliable cognitive faculties, making achieving (1) an external condition for positive epistemic status. The important question, then, is whether we can move from the premise that (1) is an external condition to the conclusion that Reid is an exernalist about epistemic justification. For reasons that I have already articulated, I think that the answer to this question is no. We have already seen that having a sound understanding is an external condition, as a believer has no control over whether or not their brain develops normally. This, however, is not enough to make Reid an externalist about justification. Access internalists can agree that having normally developed cognitive hardware is a precondition for having justified beliefs without also thinking that justification itself is external. Thus, the fact that Reid takes it as an epistemic positive when our beliefs are formed by reliable cognitive faculties does not settle the debate over whether he is an externalist about justification.

A further reason to think that we shouldn’t take (1) to describe Reid’s view of justification comes from the case that van Woudenberg makes for (1) being a condition for positive epistemic status (2013: 77–78). In his argument that (1) is an element of positive epistemic status, van Woudenberg appeals to Reid’s response to the global skeptic, that his trust in perception ‘came from the mint of Nature’ [183]. Even though this might be an effective response to skepticism, it is not clear that we should regard being supplied by nature as a normative epistemic status of belief. In defending his answer to the skeptic, Reid appeals, not to epistemic normativity, but to practical normativity: ‘I think it would not be prudent to throw off this belief […] If Nature intended to deceive me, and impose upon me by false appearances, and I, by my great cunning and profound logic, have discovered the imposture, prudence would dictate to me, in this case, even to put up with this indignity done to me’ [184]. Here, Reid defends his trust in perception by arguing that doing so is most prudent, appealing to practical reasons instead of epistemic reasons. If this is Reid’s approach to (1), though, then (1) would not be a fitting candidate for a view of epistemic justification.