Suppose We Know Things

Abstract

When contemporary philosophers discuss the nature of knowledge, or conduct debates that the nature of knowledge is relevant to, they typically treat all knowledge as propositional. However, recent introductory epistemology texts and encyclopedia entries often mention three kinds of knowledge: (i) propositional knowledge, (ii) abilities knowledge, and (iii) knowledge of things/by acquaintance. This incongruity is striking for a number of reasons, one of which is that what kinds of knowledge there are is relevant to various debates in philosophy. In this paper I focus on this point as it relates to the third kind of knowledge mentioned above – knowledge of things. I start by supposing that we have knowledge of things, and then I show how this supposition reshapes various debates in philosophy.

Philosophers don't always mean what they say. Or, at least, they don't always act like it. Sometimes they say one thing, but then proceed with their philosophical work – their papers, talks, teaching, and so on – as if the opposite were true. In this way, as in others, philosophers fail to practice what they preach.

A case in point can be found in contemporary epistemology when it comes to philosophers’ basic taxonomies of knowledge – how they divide up knowledge into its most general categories or kinds. If you just paid attention to how contemporary philosophers act – that is, how they discuss the nature of knowledge, both in print and in person, or how they conduct the myriad debates that are relevant to what knowledge is like – you'd probably think the consensus is that all knowledge (save perhaps abilities or “know-how”) is propositional. That is, you'd probably think philosophers generally agree that knowledge, whatever else it may be and however else it may be subdivided, is always constituted by beliefs in propositions.

That's how philosophers act. But that's not always what they say. Imagine that you're new to philosophy, so you don't know much about contemporary philosophical practice, and you crack open an introductory text on epistemology. You're likely to find something like the following:

The attempt to explain all the different kinds of knowledge in terms of propositional knowledge is unsuccessful. The most reasonable conclusion seems to be that there are (at least) three basic kinds of knowledge: (1) propositional knowledge, (2) acquaintance knowledge or familiarity, and (3) ability knowledge (or procedural knowledge). (Feldman 2002: 12)

Or this:

[W]e can distinguish three main kinds of ways of knowing, corresponding to three sorts of things said to be known: (1) Knowledge of facts … (2) Knowledge of a thing or person … (3) Knowledge how to do something. (Martin 2010: 1–2; also see, e.g., Audi 2010: Ch. 1; Fumerton 2006: 1)

Or you might read an encyclopedia article on epistemology that says something like this:

It is common in epistemology to distinguish among three kinds of knowledge: There's the kind of knowledge you have when it is truly said of you that you know how to do something – say, ride a bicycle. There's the kind of knowledge you have when it is truly said of you that you know a person – say, your best friend. And there's the kind of knowledge you have when it is truly said of you that you know that some fact is true – say, that the Red Sox won the 2004 World Series. (Fantl 2017; see also, e.g., Ichikawa and Steup 2017; Steup 2017)

So while contemporary philosophers act as if there's just one kind of knowledge (or maybe two), quite often, in their basic introductions to epistemology – aka, what we use to teach – they say that there are three kinds of knowledge. There's propositional knowledge, of course. And I already alluded to the controversial “know how.” But what's this third kind of knowledge – this knowledge by acquaintance/of things?

The terminology comes from Bertrand Russell (1911, 1912), who distinguishes between two kinds of knowledge: “knowledge of truths” (i.e., propositional knowledge) and “knowledge of things”, which is constituted, not by beliefs in propositions, but by awareness of objects and properties. According to Russell, this knowledge comes in two varieties: knowledge by acquaintance and knowledge by description. For Russell, this knowledge is not of truths, yet it really is knowledge.

Whether or not the above authors agree with every aspect of Russell's view, their assertion that there is knowledge of things is striking for at least three reasons. First, it's striking because these authors then go on to completely ignore knowledge of things. Their introductory texts – their surveys of epistemology – drop the subject entirely. The same goes for every other introduction to epistemology written in the last 20 years (at least that I know of). You'd think that if these philosophers agreed that there is knowledge of things and aimed to give a general survey of knowledge, then they'd discuss it some. But they don't.¹

It's also striking that knowledge of things is posited in the above texts because, in general, epistemologists don't talk about it. There are some exceptions.² But not many. You'd think that if everyone agreed that there's this other kind of knowledge, then a fair few studiers of knowledge would study it. But few have.

Finally, it's striking that epistemologists acknowledge but then set aside knowledge of things in their surveys of epistemology, because this knowledge is potentially relevant to various topics and debates that have been central to epistemology over the past few decades. Here are some of the topics I have in mind: The Given (and foundationalism more generally), the epistemic significance of experience, the problem of the speckled hen, expertise, the rationality of perception, disagreement, self-knowledge, explanation, certainty, and transformative experience. You'd think that if philosophers really believed that knowledge of things exists and that its existence is obvious enough to drop into introductions to epistemology, then they'd at least mention it – maybe even discuss its import – when relevant. But, in general, they don't.

All of this serves to reinforce my claim that, in this case, philosophers have failed to practice what they preach. I won't speculate or dwell on why epistemologists have, in general, set aside or ignored knowledge of things. Instead, in this paper I'll focus on why it matters. In particular, I'll focus on the third striking fact above – the fact that knowledge of things bears on various debates in philosophy. I'll start by supposing, as so many epistemologists do, that knowledge of things exists. Then I'll show how this supposition reshapes – in some cases, maybe even settles – a number of recent debates in philosophy.

My main aim here is to shed light on the broad and far-reaching philosophical significance of knowledge of things and to thereby encourage more reflection on the subject. So what I'll do is move briskly through a wide range of debates. After saying a bit more about what knowledge of things is supposed to be, I'll start with some debates in epistemology. Then I'll move on to debates in other areas of philosophy. There's a lot to say concerning each of these debates. And I won't be able to say the half of it. What I hope to do instead is give a glimpse of what things in philosophy might be like if, in this particular case, we practiced what we preach – that is, if we owned up to knowledge of things and incorporated it into our philosophical work.

1. Knowledge of things

Suppose we know things. That is, suppose that some of our knowledge is of things. What are we supposing?

Although the authors of the above texts agree that knowledge of things exists, they don't seem to agree on what it is. Some seem to think of it as limited to a kind of interpersonal knowledge. Others seem to think of it more in line with Russell's view – as a broader category that includes knowledge of various things.³

I prefer the latter option. Not only is Russell (1912) the source of this terminology, his general approach to knowledge of things is, I think, particularly helpful. So I'll start with his view. For Russell, there are two kinds of knowledge: knowledge of truths and knowledge of things. He says ‘knowledge’ in the former sense is:

the sense in which what we know is true, the sense which applies to our beliefs and convictions, i.e. to what are called judgments. In this sense of the word we know that something is the case. (Russell 1912: 69)

For Russell, knowledge of truths is propositional knowledge. And it is distinct from knowledge of things, which comes in two varieties: knowledge by acquaintance and knowledge by description. Russell describes acquaintance as follows:

I say that I am acquainted with an object when I have a direct cognitive relation to that object, i.e. when I am directly aware of the object itself. (Russell 1911: 108)

Russell (1911: 108) then describes acquaintance as the direct “presentation” of objects and properties to one's mind and says that, strictly speaking, we are only ever acquainted with sense data, our awareness of sense data, and a few other things.⁴ All other knowledge of things is indirect, according to Russell, and counts as knowledge by description – it's the kind of knowledge we have when we are aware of something as satisfying some description or falling under some concept, such as “the King of Jordan,” “the cashier,” or “that thing there.”

Russell maintains that knowledge by description presupposes some knowledge of truths (1912: 73). For example, in order to know of a cashier as “the cashier,” I must know certain propositions about what cashiers are. However, he also holds that some knowledge of things does not presuppose or require knowledge of truths. Take my knowledge of the color I see when looking at my coffee. I know of it as brown (which requires some background knowledge of truths), but I may also know, by acquaintance, of the very specific shade of brown that I see, for which I have no concept. This knowledge of things does not require knowledge of truths.

Some of the particulars of Russell's (1911, 1912) view are unpopular with contemporary philosophers (e.g., sense data). But we can set these shortcomings aside and focus on the essential core of his view, which is that some knowledge – i.e., knowledge of things – is constituted, not by beliefs in propositions, but by awareness of properties and/or objects. Some paradigm cases come from perception (though the details depend on your view of perception).⁵ For example, when I'm sitting in a coffee shop, and I look around, I see all sorts of things – chairs, tables, and people; colors, shapes, and spatial configurations – and my awareness of these things constitutes knowledge of (at least some of) them.⁶ The same goes for other sense modalities. Another potential source of knowledge of things is introspection. I may know of a sharp pain in my knee, a tickle on my elbow, and my thoughts about Russell's epistemology.⁷

There's plenty more to say about this Russellian account of knowledge of things (see Duncan (2020) for more). But the above brief sketch should be sufficient to give some sense of the kind of non-propositional knowledge that contemporary philosophers ignore. In giving this Russellian account, I don't mean to imply that all philosophers who accept knowledge of things are, or should be, committed to every aspect of it. It's not the aim of this paper to show that some specific view about knowledge of things is right; rather, it's to show how knowledge of things bears on various debates in philosophy. So if, at the end of the day, aspects of the above account need to be adjusted or rejected – even if the whole thing has to go – so be it. That should be part of the conversation we're not having. The point is: This knowledge, which philosophers posit but then neglect, bears on a lot of debates in philosophy. This point can be seen without assuming much about knowledge of things other than that it is constituted by awareness of properties and objects. Accordingly, the claim that I will now focus on and stress the importance of is simply:

We Know Things: Some of our knowledge is constituted, not by beliefs in propositions, but by awareness of properties and/or objects.

2. Knowledge of things in epistemology

Now that we've see what knowledge of things is (or is supposed to be), we can begin to see how it is relevant to various debates in philosophy. In this section I will discuss how the supposition that we know things – i.e., that some of our knowledge is of things – bears on a number of ongoing debates in epistemology.

2.1. The epistemic significance of experience

One ongoing debate in epistemology is about the “epistemic significance” of experience. Everyone agrees that perceptual experience plays some role in generating perceptual knowledge. But this role has proven elusive. A lot of philosophers, who seem to assume that all perceptual knowledge is propositional, say that perceptual experiences help generate perceptual knowledge by bearing certain relations to perceptual beliefs – e.g., by causing or justifying them.⁸ But these solutions are problematic. For certain cases, such as Block's (1995) “superblindsighter” – i.e., a possible case in which someone is fully “access conscious” of external things but not phenomenally conscious of them – seem to show that our perceptual beliefs could be justified without us having the relevant perceptual experiences (see Byrne 2016). So then the question remains: Why do we need perceptual experiences for perceptual knowledge? How do perceptual experiences per se help generate perceptual knowledge? These questions seem to lack a clear answer.⁹

The problem could be put in the form of an argument with an unsavory conclusion (adapted from Byrne 2016: §1, 2):

1. 1. Our perceptual beliefs would still be justified if we perceived the world without having perceptual experiences.

2. 2. If our perceptual beliefs would still be justified if we perceived the world without having perceptual experiences, then our perceptual experiences do not contribute (epistemically) to our having perceptual knowledge.

3. 3. Therefore, our perceptual experiences do not contribute (epistemically) to our having perceptual knowledge.

Almost everyone agrees that (3) is false. But (3) is entailed by (1) and (2), each of which has some plausibility. Most philosophers who address this issue deny (1). Yet their justifications for doing so have met with harsh criticisms. Thus, as of yet, we have no compelling reason to deny any of (1), (2), or, consequently, (3).

But now suppose We Know Things. That is, suppose that some of our knowledge – including some of our perceptual knowledge – is non-propositional and is rather of things. Then we have a rather straightforward answer to the above questions about how experience helps generate perceptual knowledge. That answer: It is knowledge. Perceptual experience helps generate perceptual knowledge by being it. This answer sidesteps the elusive experience-belief link. And it thus provides an elegant solution to the problem.

In terms of the above argument, this solution allows us to dodge the troublesome (1) – to grant it for the sake of argument – and instead deny (2). It allows us to say that even if our perceptual beliefs would be justified without our having perceptual experiences, our perceptual experiences still do contribute to our having perceptual knowledge, by being knowledge. So then (2) is false. As is (3) – our unsavory conclusion.

What this solution does is locate the epistemic significance of experience not in its relation to perceptual beliefs – in how it causes or provides evidence for perceptual beliefs – but rather in the experience itself. Most agree that perceptual experience has epistemic significance – epistemic oomph. But recently philosophers have assumed that this oomph comes from how experience generates something else – i.e., perceptual beliefs. But if there is knowledge of things, as many epistemologists readily admit, then we needn't look to how experience generates belief to locate that epistemic oomph. Experience itself is already chock full of it. Thus, simply owning up to knowledge of things solves a seemingly intractable problem in epistemology.¹⁰

Which isn't to say it solves every problem in the vicinity. For example, one might still wonder: What is the epistemic significance of perceptual experience for propositional knowledge – i.e., perceptual knowledge of truths? That there is epistemically significant knowledge of things does not, by itself, answer that question. However, the supposition that we know things does at least change how we can approach the question. Instead of looking for a link between perceptual experience and perceptual belief that explains the epistemic oomph of experience (which knowledge of things already explains, without the link), we can instead look to elucidate ways in which knowledge of things gives rise to knowledge of truths. This isn't an issue of where the epistemic oomph comes from – knowledge of things already has it. It's rather an issue of how that oomph transfers to knowledge of truths. Which is a different issue – a different debate.

However, although it is a different debate, it is perhaps worth saying a bit more about how knowledge of things yields propositional knowledge. One way knowledge of things may generate such knowledge is by figuring in forms of reasoning (that are sometimes overlooked) – such as association or simulation. I'll return to this point later. Another way it may do so is by rationalizing propositional knowledge more directly. In a recent paper, Kriegel (2018) draws on Brentano's logical system to show that propositional states can be paraphrased into objectual states and vice versa. This raises the possibility that knowledge of things (which is constituted by objectual states of awareness) rationalizes beliefs via a simple (and safe, reliable, rational, justified, etc.) process of paraphrase. This project is nascent, but it shows how altering assumptions around knowledge of things also alters the avenues for further discussion within epistemology.

2.2. The problem of the speckled hen

Another, related debate in epistemology concerns the problem of the speckled hen.¹¹ Suppose I see 48 speckles on a hen. I see exactly 48 speckles, and yet I am not justified in believing that there are 48 speckles on the hen or even that I see 48 speckles there. This raises a problem for philosophers, especially foundationalists, who wish to explain the epistemic potency of experience – how it generates knowledge – but who also assume that all knowledge is propositional and so explain this epistemic potency solely in terms of experiences’ relation to perceptual beliefs. Unlike the previous debate, this isn't a question of what experience is necessary for; rather, it's a question of what it is sufficient for. Some philosophers want to say that S's experiencing P is sufficient for S's having prima facie justification for believing P. That way they can explain how experience helps generate perceptual knowledge. But they cannot, or ought not, acknowledge that experiencing 48 speckles is sufficient to justify the belief that there are 48 speckles there. Hence the problem.

Here's the problem put as an argument against foundationalism (adapted from Sosa 2003: Ch. 7; Markie 2009):

1. 4. S has foundational sensory knowledge of P only if experiencing P is sufficient for prima facie justification for believing P.

2. 5. Experiencing P is not sufficient for prima facie justification for believing P.

3. 6. Therefore, S does not have foundational sensory knowledge of P.

Foundationalists deny (6). So they must deny either (4) or (5). Some deny (5). But the problem of the speckled hen makes this position tenuous. Others deny (4) by saying that foundational beliefs require more for their justification than experience – that it also requires direct awareness of the correspondence between our experiences and our foundational beliefs (e.g., Fumerton 2005). But this approach diminishes the epistemic potency of experience, and many find it too demanding an account of foundational justification.

But now suppose We Know Things. A solution emerges. Foundationalists can grant (5) (even if just for the sake of argument) and deny (4) without diminishing the epistemic potency of experience or demanding too much for foundational knowledge. They can sidestep the experience-belief link and appeal instead to knowledge of things to account for the power of experience to generate perceptual knowledge, while at the same time agreeing that I don't know that there are 48 speckles there. Specifically, they can say that I know of 48 speckles, while also denying that I know that there are 48 speckles there. Why? Because that's a different kind of knowledge – it's propositional knowledge – which I clearly lack in this case. This solution accounts for the epistemic potency of experience – it explains how perceptual experience yields a substantive epistemic grasp of fine-grained details that we experience – without also falsely implying that we have beliefs about all such details. So, in this way, foundationalists (or anyone who wants to explain the epistemic potency of experience) can satisfy one of their main objectives without falling prey to the problem of the speckled hen.

2.3. Others debates having to do with foundationalism

The two debates I just discussed are tied to other debates about foundationalism. So the fact that knowledge of things bears on the former suggests that it may bear on the latter. And, indeed, it does – very much so.

Consider two examples. Sellars (1956) famously objects to foundationalism by arguing that alleged foundational beliefs such as “This is green” (when looking at something green) couldn't be justified in a way that vindicates the idea that there is an ultimate foundation for our knowledge that doesn't itself rest on yet further knowledge. Here's one way to construe the argument (adapted from Sellars 1956; Bonjour 1978):

1. 7. Either our alleged foundational beliefs require justification or they don't.

2. 8. If our alleged foundational beliefs do require justification, then they are not really foundational, because they depend on other beliefs for their justification.

3. 9. If our alleged foundational beliefs do not require justification, then they are not really foundational, because they cannot justify other beliefs.

4. 10. Therefore, our alleged foundational beliefs are not really foundational.

5. 11. If our alleged foundational beliefs are not foundational, then foundationalism is false.

6. 12. Therefore, foundationalism is false.

This presentation of the argument makes a certain assumption explicit – namely, that our foundational knowledge is made up of beliefs. Foundationalists typically adopt this assumption – that is, they assume (11) – and then attempt to resist (8), (9), or both. However, if we have knowledge of things, then it is natural to think (as Russell did) that our most basic knowledge isn't made up of beliefs like “This is green”, which, as Sellars argues, requires for its justification background knowledge about what ‘green’ is or refers to; rather, it's made up of awareness of objects and properties (e.g., greenness), which doesn't require background knowledge about what those things are. This awareness wouldn't be conceptual or propositional, and, contrary to what Sellars and others assume, it wouldn't need to be in order for us to have foundational knowledge, because it wouldn't need to justify beliefs like “This is green” in order for us to have such knowledge – the awareness itself would already constitute knowledge. Thus, foundationalists can respond to Sellars’ objection by arguing that it depends on a false presupposition – namely, (11). As I mentioned before, foundationalists still need to show how knowledge of things yields other knowledge (perhaps in line with my earlier suggestions).¹² But having a response to Sellars’ argument is already a significant achievement.

Another worry with foundationalism is that it leads to external-world skepticism.¹³ The idea is that foundationalism offers too demanding a picture of what it takes to get from experience to justified belief and, thus, implies that we don't have much, if any, knowledge of the external world. Here it is in argument form (adapted from BonJour 2003: Ch. 10; Sosa 2003: Ch. 8):

1. 13. If foundationalists are right about what it takes to justify beliefs, then none of our beliefs about the external world are justified.

2. 14. If none of our beliefs about the external world are justified, then we have no knowledge of the external world.

3. 15. Therefore, if foundationalists are right about what it takes to justify beliefs, then we have no knowledge of the external world.

As before, this version of the argument makes explicit the (common) assumption that our foundational knowledge of the external world is made up of beliefs – i.e., propositional knowledge – and that we have to transition from experience to belief in order to know about the external world. However, if instead some our foundational knowledge is of things, then this worry doesn't arise. For then some experiences of the external world already count as knowledge. So (14) is false. There's no need to traverse a bridge from experience to belief in order to know about the external world. Hence, this skeptical worry vanishes.¹⁴

These are just two examples of how the supposition that we know things bears on debates having to do with foundationalism. No doubt there are others. Whether or not our guiding supposition has the potential to settle these debates, it is clearly relevant to them and, thus, worthy of further consideration in that light.

2.4. The rationality of perception

Siegel (2017) argues that, like beliefs, perceptual experiences can be more or less rational. They have what she calls “epistemic charge.” She argues for this conclusion by way of arguing that perceptual experiences are sometimes the product of inferences from a subject's background beliefs. This, she argues, is sufficient to give perceptual experiences rational standing. Hence, they are epistemically charged.

Here's the argument:

1. 16. Some perceptual experiences are products of inference.

2. 17. If a perceptual experience is a product of inference, then it has rational standing.

3. 18. Therefore, some perceptual experiences have rational standing.

How does knowledge of things bear on Siegel's conclusion or argument? Well, it seems to confirm her conclusion – i.e., (18). If some perceptual experiences are knowledge, then, yes, they have rational standing. On the other hand, the supposition that we know things diminishes the importance of Siegel's argument. For, if we know things, then there is a much more straightforward (perhaps even analytic) argument that goes from the premise that there is knowledge of things to the conclusion that some perceptual experiences are epistemically charged. No need to fight over Siegel's argument, which is very controversial.

This last point illustrates just how profound the incongruity can be between what philosophers say and do regarding knowledge of things. Again, Siegel's (2017) position is very controversial. But why should it be!? If it's obvious that we know things, as various philosophers suggest, then shouldn't it be obvious that (18) is true? And yet, despite what philosophers say about knowledge of things, here, as elsewhere, they act otherwise.

2.5. Other debates in epistemology?

I've now discussed several debates in epistemology upon which knowledge of things bears. There are many more. I don't have room to examine them all in detail, but here are some pregnant questions to illustrate my point:

Explanation: What counts as an explanation? A lot of philosophers seem to assume that all explanations are, or must be, propositional. But suppose I answer your request for an explanation about x by showing you a picture or map of x, or by pointing at x, thereby giving you knowledge of x. Might that count as an explanation?¹⁵

Intuitions: Are some intuitions based on knowledge of things – on what we see, hear, or feel – and are called “intuitions” in part because their bases are inexpressible?

Expertise: Is some expertise constituted by knowledge of things? Part of sommeliers’ (wine experts) expertise seems to be in their grasp – indeed knowledge – of subtleties in wine that many of us don't pick up on and that can't, in all honesty, be fully expressed with descriptions like “notes of old catcher's mitt.” This is easily explicable with knowledge of things, less so without it.

Transformative experience: Do transformative experiences depend on knowledge of things? L.A. Paul (2014), among others, argues that having certain experiences is important, even necessary, for gaining knowledge that bears on certain life-changing decisions. But if all knowledge is propositional, then it seems that we should be able to gain the relevant knowledge from a textbook or lecture, for example. On the other hand, if some knowledge (of things) is constituted by experiences, and is independent of propositional knowledge, then it's easier to see how having certain experiences is crucial for gaining certain knowledge.

These questions help illustrate the broad relevance of knowledge of things to epistemology. Clearly, the claim that we know things, which many epistemologists not only allow but insist upon, makes a huge difference to epistemology. And yet in recent decades this fact has hardly been discussed, let alone grappled with.

3. Knowledge of things beyond epistemology

And it's not just in epistemology. Knowledge of things also bears on debates in other areas of philosophy. I'll now discuss a few such debates, thus illustrating the broad philosophical significance of knowledge of things.

3.1. Philosophy of mind: the Knowledge Argument

One debate in which some philosophers actually have appealed to knowledge of things is the debate over Jackson's (1982) Knowledge Argument against physicalism. Here's a version of the argument:

1. 19. When Mary leaves her black and white room and sees something red for the first time, she learns something new.

2. 20. But Mary already knew all of the physical facts while in her room.

3. 21. Therefore, her new knowledge is not of physical facts.

4. 22. Therefore, there are non-physical facts.

5. 23. Therefore, physicalism is false.

But now suppose We Know Things. Conee (1994) and Tye (2009) appeal to this very supposition to respond to the Knowledge Argument. What they do is deny the inference from (21) to (22). They agree that Mary's new knowledge isn't of physical facts, but they deny that this implies that there are non-physical facts. For they argue that what Mary learns upon leaving her room isn't new knowledge of facts (or propositions) at all; rather, it's knowledge of things – knowledge of redness – which isn't communicated to Mary in her room. So Conee and Tye accept the plausible claim that Mary learns something new upon leaving her room, while also safeguarding the physicalist claim that all facts are physical.

There's plenty more to say here. For example, although Conee's (1994) and Tye's (2009) take on what Mary learns may be right, there's still an important question of whether, given physicalism, Mary should be able to gain all such knowledge in her room – that is, third-personally (even propositionally). For one might think that the physical is, in general, open to the empirical sciences and thus knowable third-personally. If so, then her acquisition of any new knowledge upon leaving the room still casts doubt on physicalism.

But my main point here is just that the significance of knowledge of things extends even beyond epistemology. In this case, it extends into one of the most widely discussed topics in philosophy of mind.

3.2. Philosophy of religion: suffering and transcendence

It also extends into philosophy of religion. Stump (2010) posits “Franciscan knowledge,” which she construes as non-propositional knowledge of oneself and others. It's a kind of knowledge of things – just of people. Stump appeals to Franciscan knowledge to address the problem of evil. She argues that in order to understand the suffering of others, and God's relationship to that suffering, we must appeal to this other kind of knowledge. For, on her view, it's only through awareness of people and their stories that we can begin to understand the true nature and meaning of their suffering. Whether or not Stump is right, clearly knowledge of things is potentially relevant to this debate in philosophy of religion.

The same goes for other debates in philosophy of religion. For instance, some philosophers and theologians are concerned with a tension between the idea that God is transcendent – and thus beyond human conception and description – and the idea that God is knowable. Here's an argument that expresses this tension (adapted from Keller 2018):

1. 24. God is beyond human conception.

2. 25. Therefore, we have no conceptual knowledge of God.

3. 26. All propositional knowledge is conceptual.

4. 27. All knowledge is propositional.

5. 28. Therefore, we have no knowledge of God.

Knowledge of things eases this tension. For its existence entails the falsity of (27). And it allows one to say the following: We know of God, but not that God is thus-and-so. So we have knowledge of God, but not the kind of conceptual knowledge of God that some contend is forever beyond us (cf., Keller 2018).

3.3. Logic: reasoning with knowledge of things

The last issue (or cluster of issues) that I'll consider is somewhat unique in that it applies to all areas of philosophy. For it concerns our shared methodology – how we do philosophy. Philosophers are in the business of reasoning. That's our craft – it's what we talk about, assess, and, when we're on our game, it's what we do. So it is fitting that the last topic I'll discuss is how knowledge of things is relevant to how we reason.

One way to acquire new knowledge is to reason with what you already know. You start with premises or bases that you know, then you infer some conclusion that those premises rationally support. Logicians, epistemologists, philosophers of science – indeed almost all philosophers these days – tend to assume that the bases upon which we reason from old to new knowledge are always propositional. Perhaps this is natural. We're taught to write in complete sentences, that's how we talk to each other, and that's how we put arguments up on the board (or in a paper!). So perhaps it's natural that we treat all reasoning as propositional.

But suppose We Know Things. Since knowledge forms the bases upon which we reason to new knowledge, it's only natural to wonder whether knowledge of things, supposing it exists, can or does figure in our reasoning.

One way that states of awareness (e.g., perceptual experiences) figure in our reasoning is by providing inputs to it. For example, I see the specific look on the cashier's face, and I use what I see to reason that he's having a rough day. You might think that's all, though. And you might think that these inputs just plug into propositions that we then do “real” reasoning with. But that's questionable. For, in certain cases, experiences appear to feature in our reasoning, not only as reasons, but as the very things we deliberate with and draw conclusions on the basis of. They seem particularly well suited to do this in certain types of reasoning.

Consider association, for example. When the cashier in the coffee shop I'm in responds to my questions about the menu with a certain manner and tone of voice characteristic of sarcasm, I may think about his manner and tone, and then associate these characteristics with sarcasm. Then, given the context, I may also associate that sarcasm with displeasure, and thus conclude that the cashier is displeased.¹⁶ Some of the background knowledge underlying these associations may be propositional. But when I actually use association to (knowingly) infer that the cashier is displeased, what I'm associating are not propositions – they're properties, some of which I'm aware. I associate a highly specific tone of voice that I hear and a highly specific bodily manner that I see with sarcasm, and I associate that sarcasm with displeasure. So it seems the bases of association are not always propositional. Sometimes they're things of which we are aware.¹⁷

Another form of reasoning that states of awareness seem particularly well suited to figure in is instrumental reasoning. One engages in instrumental reasoning when one identifies a non-actual state of affairs that may not be desirable in itself but that will help one achieve some goal (cf. Camp and Shupe 2017). Suppose, for example, that my goal is to have a good time at the beach. In order to do that, I have to bring the right stuff. So I identify a non-actual state of affairs – my car filled with beach stuff – that is not desirable in itself, but that will help me achieve my goal. But I need to know whether my stuff will fit. How do I figure this out? One way is via simulation (see Prinz 2002: Ch. 6; Millikan 2006: 118; Camp and Shupe 2017: 103). As I stand outside looking at the trunk of my car, I mentally simulate packing various items in my car in order to determine whether the trunk will fit everything. This simulation is based in part on my perceptual awareness of the trunk and in part on stored perceptual information that helps me imagine the sizes and shapes of the chairs, towels, and other objects that I want to fit into the trunk. Again, it sure looks like what I'm reasoning with are things I'm aware of rather than propositions.

This conclusion is also supported by well-known experiments on mental rotation. Shepard and Metzler (1971) gave subjects a pair of pictures of three-dimensional figures – composed of cubes in various arrangements – and asked subjects to report, as quickly as possible, whether the figures in these pictures were the same. In some cases, the figures were the same but one was just rotated relative to the other; in other cases, they were different. When the figures were the same, the time it took subjects to report that they were the same was proportional to the degree to which the one figure was rotated relative to the other. Shepard and Metzler concluded that subjects engaged in an internal, imagistic “mental rotation” of the figures in order to solve this problem. Shepard and others replicated these results in various different experimental designs in ways that heavily supported their conclusion.¹⁸ Now the existence of mental rotation is widely accepted by philosophers and psychologists. And this form of reasoning is based directly on one's awareness of things. It is based partly on one's perceptual awareness of the figures and partly on one's mental rotations of the figures. Again, these bases are, not propositions, but things of which one is aware.

Even purely logical reasoning may be done with knowledge of things. Consider Shin's work on diagrams and Camp's work on maps. Shin (1994) shows that Venn diagrams, for example, are governed by formal rules of inference that are sound and complete up to expressive equivalence with monadic first-order predicate logic (cf., Camp 2007: 153). And Camp (2007) shows that some maps have many of the formal features of sentences and propositions that make them well suited to feature in logical reasoning. What this illustrates is that logical reasoning can be diagrammatic or cartographic. This in itself doesn't show that we engage in logical reasoning with any other kind of representation. But it should open us up to that possibility. And if we accept that some of our knowledge is of things, then we should be open to the possibility that we can and do engage in logical reasoning with states of awareness – perhaps in a way that is analogous to how we reason with diagrams or maps. Indeed, some of the formal features of maps and diagrams do seem to have parallels in perceptual experience, such as their spatial structure and their inability to represent disjunctions and some quantificational information (e.g., universal generalizations). If these parallels can be further expanded and exploited, a logic of knowledge of things may be possible.

If so, then owning up to knowledge of things has the potential to not just reshape (or settle) particular debates in philosophy, it also has the potential to change the way we think about reasoning and, indeed, the way we reason. In other words, while owning up to knowledge of things is important in relation to other debates in philosophy, what may be even more important is how it has the potential to change how we do philosophy.

4. Conclusion

Philosophers say we know things. But they must not mean it. For that's not how they act. Not only do philosophers shun the topic both in their introductory textbooks on epistemology and in their other work, they also completely ignore knowledge of things in regards to the philosophical debates upon which it bears.

Which is a shame. For knowledge of things matters. If it exists, then a bunch of debates in philosophy should look substantially different. Indeed, philosophy itself – how it's done – should look substantially different.

One potential response to all of this is to just reject knowledge of things – to say that we weren't really serious, or that we were seriously mistaken, when we said that there's knowledge of things. If we adopt this response, then we can go back to business as usual and resume assuming that all knowledge is propositional.

Another potential response is to own up to knowledge of things. I prefer this response. This is for all sorts of reasons, one of which is drawn from this paper: If we own up to knowledge of things, then this may help us address – in some cases, even resolve – various seemingly intractable disputes and problems in philosophy.

At the very least we should give it a look. Since we've already told everyone we know things, we ought to at least make sure. Then, if it turns out we were wrong, we can, just this once, tell people: Do as we do, not as we say.¹⁹