1. SEEING IT FOR ONESELF

One of the most quoted passages in the philosophical literature is the opening lines of Aristotle's Metaphysics, which we can roughly translate as “Everyone by nature desires to know.” For example, Timothy Williamson (Reference Williamson2000) uses this quotation as the epigraph for his seminal book Knowledge and Its Limits.Footnote ¹ Williamson was clearly attracted to this remark because he is the leading exponent of the so-called knowledge first programme, which puts knowledge centre-stage in epistemology. If one digs a little deeper into what Aristotle had in mind here, however, I think a more complicated picture emerges. In particular, it seems plausible that Aristotle is not extolling the desirability of knowledge per se, but rather of a particular kind of knowledge.

Consider first the language that Aristotle uses in this remark. Although it is standard to interpret him as talking about knowledge simpliciter in this passage, the word he uses is actually quite unusual – eidenai, rather than, for example, epistêmê or gnosis – so it is clear that he has something quite specific in mind. In particular, this is a term which refers to a kind of knowledge which is essentially involved with seeing, where this is primarily perceptual, but also (as will see below) intellectual too. Indeed, that Aristotle has this specific kind of knowledge in mind is clear from how the passage continues, as he goes on to note the delight that we take from our knowledge gained from the senses, and from sight in particular. What Aristotle seems to be interested in here, and which he thus thinks all rational creatures are interested in, is a kind of knowledge that involves what we might well term seeing it for oneself.Footnote ²

Is seeing it for oneself the same as perceptual knowledge? I think not. Perceptual knowledge that p can, on most contemporary proposals anyway, be acquired in an entirely passive way, such that it expresses no intellectual desire for the truth on one's part but is simply gained in virtue of one having the right kind of perceptual faculties operating in a suitable environment. But what Aristotle has in mind, in contrast, is clearly something active, something which manifests a rational creature's intellectual desire for the truth. Seeing that p for oneself is thus a particular kind of perceptual knowledge: it is active perceptual knowledge that manifests one's intellectual desire for the truth, one's seeking out, via perception, of the truth.Footnote ³

I think one can immediately see the plausibility of the idea that seeing it for oneself, so construed, has some special value. In particular, there does seem to be something inherently better, ceteris paribus, about seeing it for oneself rather than coming to know it second-hand (or, for that matter, via mere passive perceptual take-up). Take any topic of interest. When given the choice, and assuming that no significant practical cost is involved, wouldn't one wish to know something by seeing it for oneself rather than by simply being told about it? Indeed, notice that this is not a mere descriptive point, but rather seems to reflect more substantive commitments we hold. Imagine, for example, a scientist conducting important experiments as part of a team, but who always opts, despite there being no practical cost to her, to being told how the experiments went rather than observing them for herself. Indeed, to make the case more vivid, imagine that the experiments are concerned with something momentous, like the discovery of the Higgs boson. Wouldn't we think there was something importantly lacking, from the perspective of intellectual virtue and character, about a scientist who was content to be told the results of such an experiment second-hand when she could easily see it for herself?

The crux of the matter is that it seems to be part of what it is to be intellectually virtuous, and hence to have a good intellectual character, that one has a standing desire to see things for oneself. As with the exercise of all virtues, one needs to bear in mind the different trade-offs that might be relevant in this regard. If one excessively desired always to see things for oneself, even when, say, the practical costs of doing so are enormous, then this would no longer be the manifestation of virtue but of vice. But that is of course true of all virtues, the exercise of which lies between two poles of excess and dearth.

The epistemic importance of seeing things for oneself is reflected in our epistemic architecture. No matter how strong a testimonial epistemic basis for knowing that p one has, it is almost always completely supplanted as the basis for belief once one has seen that p for oneself. Suppose our scientist is now persuaded to witness the experimental results first-hand. Isn't it plausible that after witnessing this her epistemic basis that these results have been achieved will be that she has seen that this is so for herself, rather than that she has been told that it is so? I think this tells us something important about the fundamental epistemic role that seeing it for oneself plays.

What goes for directly seeing it for oneself that p also holds, mutatis mutandis, for what we might call ‘mediated seeings’, as when one's seeing is mediated by technology. Imagine that one is part of a scientific team working in the wild, observing the unusual behaviour of some creatures in their natural habitat. Now imagine that some scientifically significant behaviour is on display. All other things being equal, it would be better for one's knowledge about this to be based on one seeing it for oneself directly rather than via binoculars. But it would also be better, all other things being equal, to know this by seeing it for oneself via binoculars rather than via the second-hand means of testimony. Relatedly, it's better to come to know about this phenomenon by directly seeing it for oneself first-hand than by watching a video recording subsequently. But it is still better to come to know this by seeing it on a video recording rather than via the second-hand means of someone's testimony. Ceteris paribus, the standing desire to see it for oneself enjoins one to see it for oneself as directly as is feasible. Moreover, the more directly one sees that p, the more likely it will be that one's seeing for oneself that p will become one's epistemic basis for believing that p.Footnote ⁴

Of course, one might dispute all of this, and argue that even if it's true that we put an epistemic premium on knowing something by seeing it for oneself, this just reflects an epistemic fetishism on our part. That is, one might argue that there is no epistemic basis for this preference for seeing things for oneself. What is important is just that one's inquiries lead to knowledge, and not whether that knowledge in addition involves first-hand experience of what is at issue. On this view – which we might term the simple knowledge account of inquiry – Aristotle is simply wrong to pick out a particular kind of knowledge as being especially desirable, since what rational subjects ought to prize is rather just knowledge simpliciter.

In what follows I am going to defend, and thereby articulate, the epistemic importance of seeing it for oneself. I'm going to proceed by first showing that the simple knowledge account of inquiry is highly problematic. As we will see, understanding why that proposal doesn't work will help us get a better handle on the epistemic importance of seeing it for oneself, including why this notion is not restricted to active perception.

2. CONTRA THE SIMPLE KNOWLEDGE ACCOUNT OF INQUIRY

For our purposes we will take the simple knowledge account of inquiry to be the thesis that the goal of inquiry is knowledge in the sense that what is both necessary and sufficient for legitimately closing a well-conducted inquiry is knowledge.Footnote ⁵ In particular, note that we are here talking of knowledge simpliciter, and not any particular kind of knowledge, such as the knowledge gained by seeing it for oneself (this is what makes it the simple knowledge account of inquiry).

Such an idea has been gaining traction in the recent literature. For example, in a prominent recent article, Alan Millar (Reference Millar2011) has invoked the knowledge first programme as set out in Williamson (Reference Williamson2000) in order to motivate a version of this thesis, and further claims that recognising the role that knowledge plays in legitimately closing inquiries helps us to understand the special value of knowledge. Millar argues that an inquiry whether p is legitimately closed only once: (i) one acquires knowledge that p, and (ii) one acquires knowledge of how one knows that p.Footnote ⁶ Moreover, Millar doesn't specify that he has a particular kind of knowledge in mind here. More recently, Christoph Kelp (Reference Kelp2014) has further defended the simple knowledge account of inquiry, but this time without the further requirement (ii) laid down by Millar.

One attraction of the simple knowledge account of inquiry – ably demonstrated by Kelp (Reference Kelp2014) – is that it avoids some of the problems that afflict accounts of the goal of inquiry which trade in lesser epistemic standings. Take, for example, the idea that truth is the goal of inquiry. One problem with this proposal is that what makes one's belief true could be completely, and yet undetectably, disconnected from the inquiry itself. Perhaps, for example, one's inquiry leads one to truly believe that p, but that the way one conducted that inquiry was so inapt for the purpose of finding out whether p that it is pure luck that one's belief in this regard is true. Imagine, say, that one tries to determine whether it will rain tomorrow by examining the tea leaves, because one falsely supposes that this is an effective way to determine such a truth, and that one happens in this way to form a true belief. Here is the crux: would we think that the agent's inquiry is legitimately closed by the lucky acquisition of this true belief? I think many would think not.Footnote ⁷

Adding that the goal of inquiry is not just true belief but justified true belief doesn't help. Consider, for example, a standard Gettier-style case, though constructed in such a way that the belief in question is the result of inquiry. Suppose, for instance, that one wishes to determine whether it will rain tomorrow, and that one seeks this information out in a perfectly reasonable fashion by consulting a reliable meteorological website. As a consequence, one acquires not just a true belief in this regard, but a justified true belief. However, now add to the case that this website was malfunctioning on this particular day, and that the weather predictions listed were in fact entirely random. Luckily, though, one happened to read one of the randomly produced weather predictions which turned out to be true. Is one's inquiry legitimately closed in such a case? Again, I think many would think not.

Moreover, in both cases the natural answer to the question of why these inquiries aren't legitimately closed is that the agents concerned didn't come to know the answer to the question they were asking as part of their inquiry. Indeed, the natural way to characterise inquiry is precisely in terms of a desire to know the answer to a certain question. Notice too that in both cases the subject may well think that she came to know the answer to the question, which is why she will no doubt be inclined to discontinue inquiring, thereby revealing an implicit commitment to the idea that knowledge is the goal of inquiry.

There is thus at least a strong prima facie case that can be made in favour of the simple knowledge account of inquiry, at least to the extent that inquiry in the relevant sense is not directed towards any epistemic standing that falls short of knowledge. Nonetheless, the view is problematic, and we can bring this point out by focussing on a particular range of cases where mere knowledge of the answer – even when supplemented with knowledge of how one knows the answer, as demanded by Millar – does not suffice to legitimately close inquiry. In particular, what is lacking in such cases is knowledge of a particular sort which is sufficient for understanding.Footnote ⁸

Imagine someone who is curious about the movements of the tides. Why are there tides at all? Why do they vary in strength at different times of the year? And so on. Accordingly, she begins to inquire into this subject matter, and in the process seeks out an expert in the field. Suppose now that this expert informs her that the movements of the tides are governed by the gravitational force that is exerted on the earth by the moon, something which varies throughout the year because the distance between the earth and the moon is variable as they move through their respective orbits. Since she is being given this information by an expert, and since she understands what she is being told (she has a broad conception of what gravity is, for example), she believes what she is told.Footnote ⁹ Given that this informant is authoritative about this subject matter, in believing what she is told she surely thereby comes to acquire knowledge of the answers to her questions. That is, she now knows why there are tides, and she knows furthermore why the tides fluctuate in strength during the course of the year.

Notice, however, that coming to know these propositions in this fashion is entirely consistent with our subject lacking any understanding of the subject matter in question. In particular, she can know that the tides are caused by the moon's gravitational pull while having no real explanatory grip on what it means for something like the moon to have a gravitational pull which could affect the earth's tides. After all, having some broad conception of what gravity is – e.g., that it's the sort of thing that keeps us firmly on the ground, such that the lack of it means that we float when in space – doesn't in itself suffice to give one a grasp of how the gravitational pull of the moon can have this specific effect on the earth's tides. Even while knowing that the movements of the tides are caused by the moon's gravitational pull, our subject might nonetheless be baffled as to how cause and effect are related.

One can thus have knowledge of why something is the case even while lacking the corresponding understanding. The question in hand for us whether merely knowing why something is the case – i.e., when one lacks the corresponding understanding – suffices to legitimately close inquiry. I suggest that it is quite clear that it doesn't. In the example just given, for instance, we have an inquiry which is generated by a subject's curiosity about why something is the case, and which results in the subject coming to know, but not understand, why this something is the case. Will the subject's curiosity be appropriately sated in such a case? Manifestly not. Indeed, one would expect our subject to continue asking questions of her informant until she gains a proper explanatory grip on how cause and effect are related; mere knowledge of the cause will not suffice.Footnote ¹⁰

Moreover, notice that this point is normative and not merely descriptive, in that our agent's curiosity ought not to be sated by mere knowledge of the answer to her question. In particular, an agent who was generally willing to regard inquiries as legitimately closed even though the relevant understanding had not been attained would be an agent lacking in intellectual virtue. Relatedly, an agent whose curiosity was sated by the acquisition of the kind of mere knowledge at issue in the case just described would not be someone with an intellectually virtuous character.

The foregoing suggests that what legitimately closes inquiry is sometimes not knowledge simpliciter, but rather the particular kind of knowledge of how cause and effect are related that is specifically involved in understanding.Footnote ¹¹ This might tempt us to offer instead what we might call the understanding account of inquiry, whereby understanding is always required to legitimately close inquiry.Footnote ¹² I think this would be a mistake, however, in that while some inquiries demand understanding in order to be legitimately closed, not all do.

Consider, for example, that one is a local police constable who has been charged by her superiors to find out whether the creature at the zebra enclosure of the local zoo is a genuine zebra rather than a cleverly disguised mule (zebra deception being a serious crime in these parts). Coming to know, via testimony from a trusted, and independent, informant, that the zebra is genuine will usually be enough to legitimately close this inquiry, at least as far as our constable's superiors are concerned. Crucially, since this inquiry is not geared towards finding an explanation of why such-and-such is the case, there is no suggestion here that understanding is required.

The upshot is that while it is only the kind of knowledge which underwrites understanding that legitimately closes some inquiries (those geared towards an explanation of why such-and-such is the case), we should not conclude from this that the overarching goal of a well-conducted inquiry is understanding, for there certainly are some kinds of inquiries which are legitimately closed by mere knowledge. The right picture of the goal of inquiry is thus what we might term the complex knowledge account of inquiry. On this view, while it is always knowledge which legitimately closes well-conducted inquiries, in some cases it is knowledge of a specific sort. It is thus a challenge for the intellectually virtuous subject to determine what is epistemically demanded of her vis-à-vis each specific inquiry.

In any case, the overarching point for our purposes is that the simple knowledge account of inquiry is wrong, and it was this view that seemed to conflict with what seems to be the inherent value of seeing it for oneself. In particular, consider again our police constable charged with determining whether the zebra at the zoo is bona fide. We just noted that testimony from a trusted independent informant would probably suffice to close this inquiry. But imagine now that this kind of deception is the sort of thing that is likely to be detectable even to the untrained eye at close quarter. And suppose that our constable is faced with the option of either asking the trusted informant or inspecting the zebra herself, with no significant practical issues bearing on her either way. Wouldn't we regard our constable as being rather remiss in her duties if she opted to close this inquiry by appeal to testimony rather than by seeing the zebra for herself?

With the simple knowledge account of inquiry off the table, we are in a position to explain this phenomenon by embracing the idea that often inquiries are better conducted if they involve active perception – of the kind involved when our constable inspects the zebra herself – rather than by appealing to second-hand knowledge. Sometimes merely knowing suffices to legitimately close inquiry, but that doesn't mean that merely knowing always suffices to legitimately close inquiry, nor does it mean that, all other things being equal, an inquiry isn't better conducted by involving active perception rather than by appeal to second-hand testimony.

3. PERCEPTUAL/INTELLECTUAL SEEING, EPISTEMIC VALUE, AND INTELLECTUAL AUTONOMY

I noted earlier that when Aristotle said that everyone desires to know he clearly had in mind a particular kind of knowing which essentially involved seeing. I also noted that the term he used in this regard – eidenai – while primarily concerned with visual perception, also had an intellectual aspect. I think we are now in a position to see what this intellectual aspect looks like in terms of the difference between mere knowledge and understanding. Put yourself in the shoes of the agent from the last section who was inquiring about the tides. In particular, consider how we would characterise her transition from merely knowing that the tides are caused by the moon's gravitational pull to gaining an understanding of why this is the case. Wouldn't this be naturally expressed in terms of seeing, in the sense of ‘ah, I see’? I submit that it is understanding of just this kind that represents the intellectual aspect of seeing that Aristotle had in mind.

I don't think that it is an accident that the kind of understanding things for oneself that is at issue here (intellectual seeing, for short) should be related to the kind of perceptually seeing it for oneself (perceptual seeing, for short) that we encountered earlier. In both cases, after all, we are talking about an active and direct grasping of the facts – they are both forms of seeing it for oneself, with one just more reflective than the other. Moreover, both perceptual seeing and intellectual seeing involve distinctive motivational states regarding one's desire for the truth, and they both involve a distinctive cognitive ownership that one takes for one's epistemic situation, in that one is no longer relying merely on the say-so of another subject. They are thus both manifestations of intellectual virtue.

Still, that they are importantly related in this fashion does not suffice to show that they are of special value such that they ought to be desirable by all rational subjects. What grounds that claim? I think there are at least three considerations that one can put forward in defence of this axiological thesis, of increasing strength and (unusually) increasing plausibility.

The first is to reiterate the point made in the last section regarding how well-conducted inquiries are often evaluated in terms of whether they lead to intellectual seeing (as opposed to mere knowing) or involve perceptual seeing (as opposed to relying on testimony). Insofar as we consider well-conducted inquiries as indicative of the specifically epistemic good – and how else would we determine the specifically epistemic good than by appeal to the goals of well-conducted inquiry? – then this gives us grounds to treat both kinds of seeing as having a specifically epistemic value.

This is quite a weak claim to make in the grand scheme of things, however. That something is valuable along a specific axis of evaluation does not entail that it is valuable simpliciter, any more than we can say that a big fly is big simpliciter.Footnote ¹³ Perhaps, that is, the specifically epistemic axis of evaluation marks out an axiological domain that is not particularly valuable?Footnote ¹⁴ Moreover, I think it is clear that Aristotle did not have such a narrow point in mind. His idea wasn't that rational creatures desire this kind of knowledge from a specifically epistemic point of view (i.e., qua good inquirers), but rather that rational creatures desire this type of knowledge – quite rightly – because it is good simpliciter. Clearly, then, it would be better to show that this kind of knowledge – knowledge by seeing it for oneself, whether perceptually or intellectually – has a value that transcends the purely epistemic.Footnote ¹⁵

I think that a plausible case can be made in this direction. Consider the idea that these perceptual and intellectual seeings essentially involve what are known as strong cognitive achievements, where strong achievements, cognitive or otherwise, are standardly thought of as being finally (i.e., non-instrumentally) valuable. An achievement is typically held to be a success that it is because of – i.e., attributable to – one's exercise of relevant agency (i.e., as opposed to being attributable to dumb luck, or someone else's agency, and so forth). A strong achievement in addition demands either the manifestation of a high level of skill or else the overcoming of a significant obstacle to success. So, for example, raising one's hand in ordinary conditions might well be an achievement of sorts – it's a success that is attributable to one's agency anyway – but it is not a strong achievement, since it doesn't meet the second condition (on either disjunct). In contrast, raising one's hand if one's arm is broken and in a cast might well count as a strong achievement.

The thought is that strong achievements have a special value in virtue of the fact that they represent cases of one's agency acting, in some substantive way, on the world. As we might put the point: a life full of successes where these successes represent achievements on one's part is a much more valuable than an equivalent life where those same successes are all down to dumb luck (or, say, the unseen intervention of helpful demon). Moreover, since strong achievements are clearly more valuable than mere achievements, better to have a life rich in strong achievements than a comparable life consisting only of mere achievements. There is thus at least a prima facie case for the idea that strong cognitive achievements are finally value.Footnote ¹⁶

Now consider perceptual seeings. Unlike merely passively knowing second-hand via testimony – or, for that matter, passively acquiring perceptual knowledge – perceptually seeing it for oneself characteristically will involve a strong cognitive achievement. While coming to know something via testimony can involve a cognitive achievement, it often doesn't, in that one's cognitive success is typically more attributable to one's informant's cognitive agency rather than one's own. But when one's belief is based on one's first-hand perception of the fact at issue, then one's cognitive success will usually be attributable to one's cognitive agency, and hence will count as one's achievement. Even so, where that perceptual knowledge is passively acquired, it won't count as a strong cognitive achievement. This is why perceptual knowledge which involves seeing it for oneself is so important. Since this is active perceptual knowledge, in that it is the result of a conscious effort on one's part which manifests one's exercise of intellectual virtue, it is thus the kind of epistemic standing that will tend to involve strong cognitive achievements.

The relationship between strong cognitive achievements and intellectual seeing is even clearer still. The kind of understanding at issue in intellectual seeing is, after all, precisely the sort of epistemic state which involves the subject consciously integrating the relevant items of information in order to construct the understanding. So construed we have an even more straightforward case of an exercise of intellectual virtue, one that involves significant levels of skill on the part of the subject and also, often, the overcoming of intellectual obstacles to understanding (understanding, as opposed to merely knowing, is often difficult after all). Understanding thus typically, if not universally, involves strong cognitive achievements.

Thus, if we grant that strong cognitive achievements are finally valuable, and also that perceptual and intellectual seeings typically involve such strong cognitive achievements, then we have an axiological basis for explaining the special value of seeing it for oneself. And note that the value in play here is not a specifically epistemic kind of value, but an entirely general goodness, of the kind that attaches to strong achievements more generally.

There is also a further – and, I think, more compelling – way of arguing for the final value of seeing it for oneself (whether perceptually or intellectually), which is in terms of how the epistemic standings involved manifest intellectual virtue. As we noted above, the desire to see it for oneself, whether perceptually or intellectually, arises out of a distinctive motivation to uncover and grasp the truth. This is one of the hallmarks of the intellectual virtues, in that they involve a love of the truth, and hence a seeking-out of the truth. But if one regards seeing it for oneself as manifesting intellectual virtue, then a straightforward route presents itself for explaining its special value. Intellectual virtues are standardly thought to be finally valuable, after all, not least because of their essential role in the virtuous life of flourishing.Footnote ¹⁷

We can add more flesh onto the bones of this suggestion by charting how seeing things for oneself relates to intellectual autonomy. Intellectual autonomy can be thought of as both an intellectual virtue and as a goal of intellectual virtue (i.e., being intellectually autonomous). In the latter sense, it is the good of being epistemically self-reliant, where this involves one taking ownership of one's epistemic position. In the former sense, it is the distinctive cognitive traits that virtuously promote this good.Footnote ¹⁸ I think it is clear that the overarching intellectual virtue at issue in seeing things for oneself, both perceptually and intellectually, is that of intellectual autonomy, and thus that they both promote the good of being intellectually autonomous. Just as perceptual seeing involves seeing it for oneself, rather than relying on another's testimony, so intellectual seeing involves understanding the phenomenon in question via one's own cognitive resources. In both cases, what is gained is a direct comprehension of things (of events, of facts etc.) that is not (ultimately, anyway) mediated through others and which therefore allows one to take a level of cognitive responsibility for one's beliefs which is unavailable if one merely epistemically relies on others.

Plausibly, then, when Aristotle said that all men desire to know, he had in mind the fact that seeing things for oneself (whether perceptually or intellectually) manifests the intellectual virtue of intellectual autonomy, and thus virtuously promotes this fundamental good of being intellectual autonomous. Given the point just noted about the final value of the intellectual virtues, and the final value of the goods that they promote – in both cases on account of their constitutive role in a virtuous life of flourishing – we thus get a way of underwriting the final value of seeing things for oneself.Footnote ^19, Footnote ²⁰

4. INTELLECTUAL AUTONOMY AND THE SOCIAL FABRIC OF KNOWLEDGE

One might be sceptical about the final value of intellectual autonomy on the grounds that such a view commits one to supposing that we shouldn't rely on the testimony of others, but should be completely epistemically self-reliable instead. I think the idea that intellectual autonomy has this implication is a mistake. In particular, we should resist the widespread idea in contemporary epistemology that a commitment to intellectual autonomy goes hand-in-hand with a commitment to reductionism with regard to the epistemology of testimony.Footnote ²¹ This latter view holds, roughly, that the epistemic standing of any testimony-based belief needs to be ‘reducible’ to evidence available to the recipient of the testimony. So, for example, I can only legitimately take your word on a certain subject matter if I have an independent epistemic basis for doing so – for example, by having personal experience of your reliability about this subject matter. Reductionism is a controversial view, and it is easy to see why, since it entails that gaining knowledge through testimony is often very difficult indeed.Footnote ²²

Notice, however, that intellectual autonomy as we are understanding this notion here is not committed to reductionism. In order to see this, recall the case that was made for the complex knowledge account of inquiry above. We argued there that an agent can come to know why something is the case via testimony without thereby gaining the corresponding understanding, but that, at least as regards a certain type of inquiry, an agent's intellectual curiosity is only properly sated when the relevant understanding is attained. In terms of intellectual autonomy, then, the idea is that the good of intellectual autonomy is in these cases served by attaining the relevant understanding rather than by attaining the knowledge which falls short of that understanding. Crucially, though, it is not being denied in this case that the agent acquires knowledge through testimony, and there is, more generally, nothing preventing us from characterising this acquisition of testimonial knowledge in anti-reductionist terms. Perhaps, for example, our agent doesn't have a very strong independent epistemic basis for believing this testimony, but the epistemic credentials of the testimony (that it is delivered by someone who is, in fact, an expert in the field) suffice to ensure that it amounts to knowledge. Treating intellectual autonomy as a fundamental good thus does not prevent us from holding that agents might gain testimonial knowledge while nonetheless lacking an adequate (from the perspective of reductionism) independent epistemic basis for the belief so formed.

There is a related point that needs to be made here too, which is that this conception of the good of intellectual autonomy is entirely compatible with treating the acquisition of knowledge as for the most part a social enterprise. In the case given above regarding an agent's curiosity about the movements of the tides, for example, the transition from mere knowledge to understanding will standardly be attained not via some heroic attempt on the part of the agent to work out the explanatory details entirely by herself, but rather by asking the expert further questions. Intellectual autonomy as we are characterising it here is thus not wedded to a kind of epistemic individualism which eschews any epistemic dependency on others. Instead, it is entirely compatible with epistemic dependency on others, particularly when it comes to experts. The point is that in seeking understanding one is gaining an intellectual grip on things which is one's own – i.e., which one can take epistemic responsibility for – even though the means by which this intellectual grip is attained might be highly epistemically dependent on others.

Note that this point isn't confined to understanding, but also applies to perceptually seeing things for oneself. After all, others can help one to see things for oneself more clearly, such as by directing one's attention towards salient features of the scene. The manifestation of the desire for intellectual autonomy that we find in the urge to perceptually see things for ourselves is thus compatible with there being an important social dimension in the acquisition of knowledge, just as it is when it comes to intellectual seeing.Footnote ²³

5. CONCLUDING REMARKS

My goal has been to articulate the value of seeing it for oneself, where this is revealed to have both a primary perceptual aspect and a secondary intellectual aspect. As we have seen, this value is ultimately rooted in the role that seeing it for oneself plays in the promotion of intellectual autonomy and, thereby, a virtuous life of flourishing. Along the way we have also seen the implausibility of a certain view about the goal of inquiry, which casts that goal in terms of knowledge in general, rather than recognising that often it is a particular kind of knowledge that is (rightly) sought. Finally, we have seen that tying the good of seeing it for oneself to the goodness of intellectual autonomy is entirely compatible with the idea that there is an important social dimension to our epistemic practices.Footnote ²⁴