Introduction: The Error of Confusing Intuition with Knowledge
Moritz Schlick's work shaped logical empiricism and thereby an important part of philosophy in the first half of the twentieth century. A continuous thread that runs through his work is a philosophical diagnosis of, as he put it, the ‘great errors in philosophy’. One can extract from Schlick's writings a list of these ‘great errors’, together with the people who committed them:
Descartes: The Cogito argument is a fallacy. There is no privileged self-knowledge. (See Schlick Reference Schlick1925: 84–85 [78])Footnote 1
Kant: According to Kant, intuitions are necessary for knowledge. ‘But in truth intuition gives us no knowledge whatever; it is wholly inessential for this purpose. It provides, to be sure, an acquaintance with objects, but never knowledge of them’. (See Schlick Reference Schlick1918: 282 [203–204])
Brentano: Brentano (see Reference Brentano, Rancurello, Terrell, McAlister and Kraus1924: 111 [I: 203]) is wrong to hold that we have immediate perceptual knowledge of our present mental life. (See Schlick Reference Schlick1925: 84f. [80])
Husserl: Husserl (see Reference Husserl and Gibson1913: 13 [12f.]) is mistaken when he claims that there are intuitions of essences that provide a foundation for philosophy as a rigorous science. (See Schlick Reference Schlick1925: 82 [76])
Russell: We are acquainted with logical atoms that form the end of logical analysis. But while there is acquaintance, there is no knowledge by acquaintance (see Schlick Reference Schlick, Friedl and Rutte1932: 218, among others).
Mach/ Avenarius: The fundamental objects of the world are neutral elements. These are supposed to be known by acquaintance. (See Schlick Reference Schlick1925: 231 [212–13])
I am not sure whether these are all the great errors of (theoretical) philosophy, but the above is an impressive list of important philosophical views that Schlick takes to be erroneous. The single source of all these great errors consists in thinking that we possess a particular kind of intuitive knowledge—intuitive knowledge of a thinking substance, particulars in space and time, mental activities, essences, and so on:
In the ranking of immediate intuition as knowledge lies the source of the great errors and mistakes (even in Kant, for example). The consequences of this error extend deep into the practice of the individual sciences. (Schlick Reference Schlick, Mulder, Kox, Hegselmann, Mulder, Kox, Hegselmann and Heath1933–34: 72 [105])
In his General Theory of Knowledge, Schlick expounded the error in Russellian terminology:
Here we uncover the great error committed by the philosophy of intuition: the confusing of acquaintance [Kennen] with knowledge [Erkennen]. (Schlick Reference Schlick1925: 83 [77])
Why do philosophers confuse intuition or acquaintance with knowledge? Some concepts are acquired by composing concepts one already possesses. But not all concepts can be acquired in this way on pain of infinite regress. The acquisition of some concepts must be possible without already possessing conceptual ‘building blocks’. How? Schlick outlined a tempting answer:
When I gaze at a red surface, the red is part of the content of my consciousness; I experience it, and only in this experience of immediate intuition, never through concepts, can I know what red is. … Are we not then fully justified in saying that intuition is knowledge? (Schlick Reference Schlick1925: 81 [76]; my emphasis.)
Acquaintance is preconceptual and noninferential and, in turn, allows for the acquisition of concepts. If hearing the note F is coming to know F, this knowledge can ground possessing a concept of F: If one has preserved knowledge of F, one has a redeployable concept of F.
When the confusion of acquaintance and knowledge is uncovered, we come to see that there is a sharp dichotomy between intuition and knowledge, and the two play opposed roles in our life and science. Schlick sums up the dichotomy in the following diagram (Schlick Reference Schlick, Friedl and Rutte1932: 226):
The dichotomy Schlick set up has been highly influential. For example, Friedman writes:
I believe that this ‘dialectical opposition’—exemplified so clearly and painfully in Schlick's struggles with the form/content distinction—animates both logical positivism and the analytic tradition as a whole. In Frege and Wittgenstein—thinkers whose primary concern is with mathematics and logic—the formal element predominates completely, and concern for empirical knowledge and epistemology in the traditional sense shrinks to the point of vanishing. In Schlick and Carnap—thinkers with an equal concern for physics and empirical science—the two elements are brought together with clearly evident resulting strains and tensions. … I believe that the ‘dialectical opposition’ between logic and experience, form and content, also informs Carnap's later work and the post-positivist philosophies of Quine and the later Wittgenstein. (Friedman Reference Friedman1983: 514)
The philosophical stakes are high. So why does Schlick deny that acquaintance (immediate intuition) is knowledge? Is acquaintance indeed fundamentally different from knowledge and if it is, what kind of mental state is it?
As far as I can see these questions have not yet been answered in the literature. Friedman (Reference Friedman1983: 508) points out that Schlick's diagnosis that all metaphysics is based on a confusion between intuition and knowledge is ‘startling’, but he does not go on to explore Schlick's arguments in detail. Shelton (Reference Shelton1989: 315) mentions Schlick's insistence that acquaintance is not knowledge, but does not try to justify it.
I will answer these open questions by reconstructing Schlick's arguments. Their guiding assumptions will become clear if we attend to the influence on Schlick's work of Schopenhauer on the one hand and the economy philosophy of Avenarius and Mach on the other.Footnote 2 While Schlick finds much to criticize in the particular formulations of Mach and Avenarius, his view of knowledge elaborates, as he himself acknowledges, the core theses of these philosophers. Working through Schlick's arguments for the view that acquaintance is not knowledge will reveal core assumptions of economy philosophy and give us a perspective to assess them.
One of Schlick's main targets is Russell's view of knowledge by acquaintance. I will therefore prepare the further discussion in the next section by expounding Russell's view. I will then go on to develop in detail his arguments for the conclusion that acquaintance neither is nor can constitute knowledge.
1. Getting a Target in View: Russell on Knowledge of Things
For Schlick and his followers, Russell's theory of knowledge by acquaintance is a representative example of a theory that confuses experience with knowledge. For example, Schlick's students Blumenberg and Feigl (Reference Blumenberg and Feigl1931: 285) take Russell's distinction between knowledge of things and truth to be one between knowledge proper and immediate experience, not knowledge by acquaintance (see also Friedman Reference Friedman1983: 500; for a recent development of Russellian acquaintance as an epistemic relation see Campbell Reference Campbell, Beckermann, McLaughlin and Walter2011). In Problems of Philosophy Russell develops a distinction between two kinds of knowledge of things: knowledge by acquaintance and knowledge by description. Knowledge of things by description requires knowledge of truths ‘as its source and ground’ (Russell Reference Russell1912: 25). One knows a by description if, and only if, one knows (the truth) that there is exactly one F and a is the F.
In contrast, knowledge of a thing by acquaintance ‘is essentially simpler than any knowledge of truths, and logically independent of knowledge of truths, though it would be rash to assume that human beings ever, in fact, have acquaintance with things without at the same time knowing some truth about them’ (Russell Reference Russell1912: 25). Instead of the technical term ‘acquaintance’ Russell uses also the nontechnical expression ‘direct awareness of’ (see Russell Reference Russell1912: 26). When you see a color, you have direct awareness of it; when you hear a note, you have direct awareness of it. Awareness of x is direct if it is not dependent on awareness of something else.
It is logically possible that a thinker S knows a by acquaintance without there being a truth about a that S knows. For instance, S can be conscious of her current pain and thereby know it without knowing any truth about it. However, under normal circumstances human beings cannot fail to acquire knowledge of some truths when and in virtue of having acquaintance with an object. If I know of my current pain and possess the concept of existence, I also know a truth about it, namely, that it exists (see Russell Reference Russell1912: 23). While we come to know truths about objects in virtue of knowing these objects, we do not come to be acquainted with objects in virtue of knowing truths about them. I can know propositions that are made true by my current pain without thereby being acquainted with it.
All mental acts that constitute direct awareness of objects have a property that mirrors the property of factivity for propositional attitudes. A factive attitude is a propositional attitude ϕ such that one can infer from ‘S ϕ-s that p’ that it is the case that p (see Williamson Reference Williamson2000: 34). Now seeing a, hearing a, being conscious of a are not propositional mental states (I follow Montague (Reference Montague2007) and others in taking there to be irreducible nonpropositional attitudes). Hence, they cannot be factive in the sense that being in them guarantees their truth. But they can have a property very much like factivity. I will call a nonpropositional attitude ρ objective if, and only if, one can infer from ‘S ρ-s a’ that a exists. In short: one can only have an objective attitude if the object it is directed at exists. For example, if S hears the note F at t, F occurs at t.
Because the awareness constituting acts implies the existence of their objects, Russell takes them to be relations between minds and these objects:
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. When I speak of a cognitive relation here, I do not mean the sort of relation which constitutes judgment, but the sort which constitutes presentation. (Russell Reference Russell1911–12: 108)
Prima facie, the cognitive relation of direct awareness is different from judgment in that the former can hold between minds and particulars and universals, while judgment holds only between minds and propositions. But by the time Russell developed his theory of acquaintance he had given up on propositions. In a paper he suggests by way of response ‘that presentation (or acquaintance) is a two-term relation of a subject, or better an act, to a single (simple or complex) object, while judgment is a multiple relation of a subject or act to the several objects concerned in the judgment’ (Russell Reference Russell1913: 76). However, he also describes seeing a table as simultaneous acquaintance with the sense-data that make up its appearance (see Russell Reference Russell1912: 25). Some of these sense-data—color and shape are likely candidates—are such that we are jointly immediately conscious of them together. So some knowledge of things is a multiple relation to several things. But even if the adicity of direct awareness may be greater than two, its minimal adicity is still two. In contrast a mind needs to connect at least a universal and another thing to make a judgment: its minimal adicity is three.
Schlick agrees with Russell that acquaintance is a brute two-place relation between a mind and a thing, either a universal or particular. But he denies that this relation is or can constitute knowledge of things. He presents different arguments for this conclusion. One line of argument appeals to our ordinary understanding of ‘know’; another is based on the metaphysics of knowledge. I will start with the first argument because a discussion of it will highlight the distinctive character of the second argument.
2. Schlick's Linguistic Argument
Russell (Reference Russell1912: 23) pointed out that several languages have different words for propositional knowledge (wissen, savoire) and acquaintance (kennen, connaître). Schlick takes this linguistic observation to suggest that acquaintance and knowledge are different concepts:
Acquaintance and knowledge are such fundamentally different concepts that even ordinary discourse has two different words for them. And yet they are hopelessly conflated by the majority of philosophers, with only too few laudable exceptions. (Schlick Reference Schlick1925: 83 [77])
He targets Russell specifically when he elaborates this point later:
Mr. Bertrand Russell distinguishes between ‘knowledge by acquaintance’ and ‘knowledge by description’ but why should the first be called ‘knowledge’ at all? The word ‘acquaintance’ alone seems to me sufficient, and then we can emphasize the distinction between acquaintance and knowledge. There is no similarity of meaning between the two. (Schlick Reference Schlick, Friedl and Rutte1932: 218; last emphasis mine)
Acquaintance and knowledge are not only different concepts; they are not even similar and a fortiori not similar enough that acquaintance qualifies as a kind of knowledge. Why? We say ‘S knows that p’ etc. This suggests that knowledge is a propositional attitude; an attitude to a proposition referred to by a ‘that’ clause. In turn, ‘S knows x’ constructions seem neither to report propositional attitudes nor to imply propositional knowledge. So if one takes acquaintance with x to be a kind of knowledge, one trades on an ambiguity (see Sellars [Reference Sellars1956: 12–25] for a similar diagnosis). Or so the Linguistic Argument against acquaintance as a form of knowledge contends.
A different way to work this point is to deny that ‘S knows x’ constructions are nonpropositional. For instance, Crane asks:
What does it mean, to know a shade of red? It might mean: to know which shade it is; or to know that it is called ‘vermillion’, for example; or know simply that it is a shade of red. (Crane Reference Crane2012: 195)
Here ‘know + direct object’ plausibly ascribes propositional knowledge. In other cases we might use these constructions to ascribe a recognitional ability or an ability that presupposes and/or yields propositional knowledge (see Craig Reference Craig1990: 148 and also Crane Reference Crane2012: 193ff. and Proops Reference Proops2014: 808). Thus, either Russell trades on an ambiguity of ‘know + direct object’ constructions or these constructions do not lend themselves to support Russell's view that there is knowledge of things.
If ordinary usage were all we could go by, Schlick would have a strong argument against acquaintance as an epistemic relation. However, there is a different justification of the view that acquaintance is form of knowledge available that does not rely on the ordinary usage of ‘knows + direct object’. It has been argued that every factive attitude implies propositional knowledge (see Williamson Reference Williamson2000: 37–38 who builds on Unger Reference Unger1975: 158ff.). Consider some examples for illustration and support:
If S sees that it is raining, S knows that it is raining.
If S remembers that it was raining yesterday, S knows that it was raining yesterday.
And so on.
There are prima facie counterexamples, but there are also good responses to the counterexamples (see Williamson Reference Williamson2000: 38). The thesis that factive attitudes imply knowledge seems defensible. Williamson takes these examples to suggest that knowledge is the most general factive mental state.
If knowledge is the most general factive mental state, direct awareness is the most general objective mental state in the sense explained in section 1. Consider:
If I see x, I am directly aware of x.
If I hear x, I am directly aware of x.
If I feel x, I am directly aware of x
And so on.
Seeing, hearing, feeling are particular ways to become directly aware of something. Russell's ‘acquaintance’ is a technical term for this most general objective attitude.
Now we can answer the Linguistic Argument. Williamson observes:
Factive mental states are important to us as states whose essence includes a matching between mind and world: and knowing is important to us as the most general factive stative attitude. (Williamson Reference Williamson2000: 40)
Russell draws our attention to the fact that such matching between mind and world can hold between a nonpropositional attitude and an object. What acquaintance, or direct awareness, and propositional knowledge have in common is that they guarantee the existence of what they are directed at. If attitudes whose essence includes a matching between mind and world deserve the title ‘knowledge’, awareness is a form of knowledge. There is an important conceptual similarity between propositional knowledge and knowledge of things, but we cannot uncover it merely by scrutinizing the use of ‘knows’.
In sum: we can respond to the Linguistic Argument in the spirit of Russell's approach. But unlike many present-day critics of Russell, Schlick does not rest his case against knowledge of things on the linguistic argument. The next section is devoted to Schlick's main argument.
3. The Argument from the Metaphysics of the Knowledge Relation
Russell argues that awareness-of has a minimal adicity of two and that this relation is a case of knowledge. Schlick disagrees. In his writings he does not use the notion of a proposition. Instead of talking of knowing truths, Schlick takes knowledge necessarily to be to be a relation between the mind and at least two things. He uses a knowledge attribution to make this plausible:
When I see a large bird in the distance and say: ‘I know that's a stork’—what do I mean by this? Obviously that the bird is of a type familiar to me, with which I am acquainted by observation or instruction, and which I have learned to designate by the name of storks. I rediscover in this bird certain features—large beaks, long legs, etc.—which are nothing new to me, and which I have been accustomed to regard as features of the stork species. The knowing therefore comes about only by comparison of a new object with an old one, and a rediscovery of the latter in the former. Knowing implies two terms: something known, and that as which it is known [Zum Erkennen gehören zwei Glieder: etwas, das erkannt wird, und dasjenige, als was es erkannt wird.] (Schlick Reference Schlick1913: 144 [477]; my emphasis)
The thesis that cognition requires two terms is a constant in Schlick's work. He repeats the ‘knowledge requires two terms’ slogan for example in his Problems of Philosophy in their Interconnection (Schlick (Reference Schlick, Mulder, Kox, Hegselmann, Mulder, Kox, Hegselmann and Heath1933–34: 96 [67]). Incidentally, Russell counts the subject as one relatum of acquaintance, while Schlick does not count the subject as a further relatum of cognition. Hence, while both talk of two terms, they nonetheless disagree about the adicity of cognition.
According to Schlick, knowledge requires two terms because knowledge is constituted by comparing one thing with something already known. The importance of comparing was stressed before Schlick by Mach:
Memory is always prepared to supply known facts which are similar to the new one, that is, share certain marks with it, for comparison and enables in this way the elementary judgment which is soon followed by the verbalized one.
The comparison, in enabling us to communicate in the first place, is at the same time the most powerful life element of science. The zoologist sees in the bones of the flying membrane of the bat, fingers. (Mach Reference Mach1894: 238–39 [266]; my translation)
Mach takes the most basic nonperceptual mental act to be comparing. It is done instinctively: we see in a particular its kind, in a present event, a future event of a certain kind, and hear in a sound or sign, facts. Schlick describes instinctive comparing as association of an idea of an object with a memory idea or a fusion with such an idea (see Schlick Reference Schlick1925: 16 [15]). Other forms of comparing, for instance, scientific knowledge, require more complex abilities (See Schlick Reference Schlick1925: 94–95 [87–88] for the thesis that there is a continuity between instinctive and scientific knowledge). But these complications are not important for our purposes.
Thus, we have an initial reason to say that there are cases of knowledge that have two terms. We will, however, need a further argument to show that it is necessary or even essential that knowledge has two terms. I will come back to this in section 6.
Schlick adds that comparing requires concepts. Therefore, knowledge is ‘by its nature conceptual’ (Schlick Reference Schlick1913: 149 [484]). This additional claim needs further justification. For instance, prototypes, paradigms, or properties may be things that serve as the additional term in the knowledge relation. I will not discuss the plausibility of Schlick's addition here and instead focus on the core argument.
Why cannot acquaintance, in general, have two terms? Because acquaintance is ‘repugnant’ to comparing and awareness of resemblance:
Intuition, as we continue to find, is the very opposite of knowledge. In pure intuition, raw contemplation, everything is utterly individual, for itself, compared to nothing. The multifariousness of experience is infinite—the same thing never recurs in it exactly as before. To abandon oneself to intuition is therefore to ignore all resemblances, to reject all combinations and order, in short to disdain everything that actually constitutes knowledge. (Schlick Reference Schlick1913: 150 [486])
I will provide clarification of this quotation in due course. But, as a first stab at it, we can reconstruct Schlick's argument now as follows:
The Argument from the Metaphysics of Cognition
(P1) Knowledge requires comparing one thing to another already known.
Therefore: (C1) Knowledge has at least two terms.
(P2) Intuition or acquaintance of x requires ignorance of resemblances of x with other things.
Therefore: (C2) Intuition or acquaintance of x can only have one term.
Therefore: (C3) Intuition or acquaintance cannot be knowledge. (From (C1) and (C2))
The premises of Schlick's Argument from the Metaphysics of Cognition are not obvious. Why believe them? The grounds on which he asserts (P2) can only be assessed after some historical scene setting. I will do this in the next two sections. I will then turn to Schlick's independent argument for (P1) to complete his line of thought.
4. Schlick and Schopenhauer on Acquaintance as Immersion
Why does acquaintance with a rule out comparing a (associating a with further things) and thereby make acquaintance essentially one-termed? In ‘Form and Content’, Schlick gives an example of an intuition of a color that sheds light on his reasoning here:
When I look at the blue sky and lose myself in the contemplation of it without thinking that I am enjoying the blue, I am in a state of pure intuition, the blue fills my mind completely, they have become one, it is the kind of union of which the mystic dreams. Bergsonian intuition is the mystical conception of knowledge. (Schlick Reference Schlick, Friedl and Rutte1932: 223)
If ‘blue fills my mind completely’, I do not compare or associate it with other sensory qualities. Hence, this remark takes us closer to the second premise of the Argument from the Metaphysics of Cognition.
Now it seems excessive to say that when I am acquainted with blue, I and blue enter a ‘mystic union’. Why does Schlick frame his point this way?
Schlick speaks of ‘Bergsonian intuition’. Bergson (Reference Bergson and Hulme1912: 43) wrote that in intuition ‘one places oneself into’ the object intuited. However, Bergson does not explain this any further, and Schlick in his papers on intuitive knowledge takes for granted that his readers do not need further explanation about the notion of acquaintance under discussion. We can make progress in understanding the conception in play here if we go back to Schopenhauer. For Schlick's remarks on acquaintance are not only influenced by Bergson, but mainly and fundamentally by Schopenhauer's theory of ‘extra-ordinary cognition’. Schlick already discussed Schopenhauer in his Lebensweisheit in 1908. He went on to lecture on Nietzsche and Schopenhauer in Rostock and Vienna (see Stadler Reference Stadler2001: 726). While Schlick rejects many Schopenhauerian theses, Schopenhauer's distinction between ordinary and nonordinary cognition is congenial to the basic tenets of the economy philosophy of Mach and Avenarius, and Schlick draws on Schopenhauer in his theory of acquaintance (Bergson is in these points so close to Schopenhauer that Jacoby (Reference Jacoby1912: 609) raised the question of whether Bergson plagiarized Schopenhauer; Jakoby went on to answer his question negatively). If we introduce the Schopenhauerian background of Schlick (and Bergson), the ‘fusion’ view of acquaintance will become clearer. So let us first get those Schopenhauerian ideas in view that Schlick will use.
First, ordinary cognition. Schopenhauer (Reference Schopenhauer, Norman, Welchman and Janaway2010: 200) holds that ordinary cognition is only an ‘instrument of the will’ (on this see Schlick Reference Schlick and Iven2013: 425–26). The ‘will’ is our drive to live and reproduce (see Janaway Reference Janaway1989: 7f.). One way to make Schopenhauer's view plausible is to use the notion of affordance; Young (Reference Young2005: 109) speaks of potential threats or allurements to the will (on perceiving affordances see Prosser Reference Prosser2011: 481). We perceive the onrushing car as a danger to us because it comes toward us, the cake as a treat for us, the water as drinkable by us, and so on. What we perceive are objects having dispositional properties that are of interest to us. Awareness of these relational properties allows the perceiver to predict future events and thereby enables her to satisfy her desires. The drinkable stuff, for example, will quench her thirst when she drinks it. Science is just the continuation of ordinary cognition:
Sciences also view things in essentially the same way, namely as nothing other than relations, connections of time, of space, the causes of natural alterations, the comparison of shapes, motives for events, and thus simply relations. (Schopenhauer Reference Schopenhauer, Norman, Welchman and Janaway2010: 199)
When reading Schopenhauer, Schlick must have been struck with the similarity between Schopenhauer's will and the drives postulated by philosophers like Avenarius and Mach. According to Mach, knowledge results from ‘an instinctive habit of mimicking and forecasting facts in thought’ that serves our welfare. If we have a habit, for example, to move from a perception to an expectation, we have knowledge of causality:
A knowledge of causality in this form certainly reaches far below the level of Schopenhauer's pet dog, to whom it was ascribed. It probably exists in the whole animal world, and confirms that great thinker's statement regarding the will created the intellect for its purposes. (Mach Reference Mach1882: 188–89 [218–19]; my emphasis)
Mach takes his economy theory of cognition to confirm a central tenet of Schopenhauer's philosophy: the drive to life, also known as the will, comes first, and we acquire knowledge to satisfy it. The drive to save intellectual effort is a particular species of the more general drive Schopenhauer writes about.
Second, extraordinary cognition. According to Schopenhauer, we have the mental strength to resist the will. If we do so, we are able to have extraordinary cognitions. Schopenhauer described what happens when we free our perceiving from the will in evocative language. Schlick (Reference Schlick and Iven2013: 425f.) quotes the relevant sentence almost completely. So will I now:
When elevated by strength of mind to stop viewing things in the ordinary way, no longer led by the forms of the principle of sufficient reason to pursue merely the relations between things (which in the end always aims at their relation to our own will), if we stop considering the Where, When, Why and Wherefore of things but simply and exclusively consider the What, if we do not allow our consciousness to become engrossed by abstract thinking, concepts of reason; but if, instead of all this, we devote the entire power of our mind to intuition and immerse ourselves in this entirely, letting the whole of consciousness be filled with peaceful contemplation of the natural object that is directly present, a landscape, a tree, a cliff, a building, or whatever it might be, and, according to a suggestive figure of speech, we lose ourselves in this object completely, i.e. we forget our individuality, our will, and continue to exist only as pure subject, the clear mirror of the object, so that it is as if the object existed on its own, without anyone to perceive it, and we can no longer separate the intuited from the intuition as the two have become one, and the whole of consciousness is completely filled and engrossed by a single intuitive image – if, therefore, the object is in this manner removed from any relation to things outside of itself, and the subject is removed from any relation to the will, then what we thus cognize is no longer the individual thing as such, but rather the Idea, the eternal form, the immediate objecthood of the will at this level: and this is precisely how someone gripped by this intuition is at the same time no longer an individual: the individual has lost himself in this very intuition: rather, he is the pure, will-less, painless, timeless subject of cognition. (Schopenhauer Reference Schopenhauer, Norman, Welchman and Janaway2010: 200)
In essence, Schlick's theory of acquaintance adopts the psychological parts of Schopenhauer's description. In order to be acquainted with an object one must overcome the will. Let us therefore look at Schopenhauer's analysis of will-free perception more closely.
Consider an example for illustration. Imagine you look at a strikingly beautiful thing; say you see the Watzmann (a mountain in the Bavarian Alps near Berchtesgaden) in the morning sun. Its beauty overwhelms and intrigues you; you devote all your attention to appreciating it. When you thus ‘lose yourself in the object’, you no longer perceive only what it affords you. When you have overcome the will and are fully absorbed, you can perceive what the object is independently of what it affords to you; you can ‘simply and exclusively consider the What’, that is, know the essence of the object. The Watzmann no longer seems to you to be an obstacle to your journey, a climbable object, and so on, but it appears as the monumental mountain it is. If we overcome the will, our perception no longer just reveals relational dispositional properties to us. This is part of what it means that ‘we lose ourselves in the object’. If I give my mind completely to the object and thereby stop to be a separate agent with its own aims, I end up as a ‘mirror of the object’ (see also Reginster Reference Reginster2008: 264).
Schlick described acquaintance in the same way and with the same words Schopenhauer described will-free perception, here aesthetic appreciation: the perceiver ‘loses himself in’ (‘sich verlieren’) the object of perception (see Schlick Reference Schlick and Iven2013: 425–26). The state I am in when I lose myself in an object merits also Schlick's description of acquaintance as a union of object and perceiver (see the quotation above). He gives further descriptions: ‘The object is no relation to something external to it, the subject external the will’ (Schlick Reference Schlick and Iven2013: 426; my translation). When I am immersed in the object, my habits of association and the drive to compare the object to others is blocked. I have overcome ‘the will’, and my cognition is no longer an instrument of it. The perceiver is ‘no longer concerned with relations following the principle of sufficient reason but instead resting and becoming absorbed in a steady contemplation of the object presented, aside from its inter-connections with any other object’ (Schopenhauer Reference Schopenhauer, Norman, Welchman and Janaway2010: 200–201). If we manage to overcome the will and immerse ourselves in an object, we are acquainted with it. But acquaintance does not make us aware of affordances and similarities. We are aware of the object and nothing else: we do not relate it to us or to other objects. Hence, what we arrive at is the feature of acquaintance that is important for Schlick's argument. To repeat: ‘In pure intuition, raw contemplation, everything is utterly individual, for itself, compared to nothing’ (Schlick Reference Schlick1913: 150 [486]).
The consequence for the metaphysics of acquaintance is that acquaintance has just one term: the object. With this in mind we can now go back to the Argument from the Metaphysics of Cognition. If we are acquainted with an object, we immerse ourselves fully in it and are not motivated to follow associations or make comparisons. This makes premise (P2) of the Argument from the Metaphysics of Cognition plausible.
To anticipate: If we manage to immerse ourselves in an object, Schlick argues, we do not come to know its intrinsic, nonrelational properties. Precisely because we have overcome the natural drive to see in an object features already experienced, we do not know it. What Schopenhauer calls ‘cognition of the form of the object’ is according to Schlick (Reference Schlick and Iven2013: 426) not cognition or knowledge at all. But if immersion in an object is not knowledge, what is it?
5. Schlick's Evaluative Conception of Acquaintance
Schlick's answer to this question is contained in his Lebensweisheit: Versuch einer Glückseligkeitslehre (1908). He argues that we can relate to an object at a time either theoretically or practically:
Man can take two standpoints in relation to the objects of the world: either he studies their nature [Wesen] or he asks for their value; he has, as Kant would express it, a theoretical or a practical interest in things. The first of these ways of considering things is the one of science and only of science, the latter is the one of all other areas of life. … Only scientific statements are not of a practical nature; all other judgements, every thought, uttered or inscribed sentence is a value judgement. (Schlick Reference Schlick1908: 86; my translation)
Implicitly, Schlick assumes that one cannot at the same time take both standpoints with respect to the same object. We take the practical standpoint toward an object if we evaluate it. We can evaluate it by making proper judgments about it— judging that this object is good—or by taking pleasure or displeasure in it. For Schlick, an object is valuable if, and only if, it is a potential source of pleasure. Pleasure makes for differences in value (see Schlick Reference Schlick1908: 86).
Under which conditions do we feel pleasure? Schlick (Reference Schlick1908: 90) endorsed the Platonic idea that one feels pleasure if a drive is satisfied. We have different basic drives and sometimes their satisfaction requires us to act. I need to buy and cook food to sate my hunger. However, often no such activity is required:
The impacts of the external world can happen in such a favourable way that our drives are satisfied without noticeable activity on our side and we only need to offer up our chalices to the showers of pleasure which flow from the heaven of coincidence. (Schlick Reference Schlick1908: 91; my translation)
Which drives are satisfied in this passive way? We do not only perceive, we have a general drive for free exercise of our senses. Schlick gives a poetic description of the passive satisfaction of this drive:
All senses are occupied in bustling activity to receive the external impressions, they feel the caress of the wind, the glowing of the landscape, the whispering of the treetops and the rumbling of the surf as an immeasurable pleasure—for certainly the drives of man are not the weakest which long for a free activity in conformity to their nature; the eye thirsts for colors and shapes, the ear for harmonies, the skin for soft caresses; and the pleasure drawn from the calm acting of the senses alone can make us happy if it is not deafened by the violence of vehement acting. (Schlick Reference Schlick1908: 91–92; my translation)
The drive for the free exercise of our senses is satisfied in perceiving, and its satisfaction is a sensory pleasure. Schlick's description of the perceptual drives seems incomplete. Is the drive to perceive satisfied if we perceive at all? Or do our senses ‘thirst’ to perceive particular shapes, forms, and sounds? It seems plausible that the general and specific drives come as a package deal.
Now we have a starting point for a positive description of acquaintance: My eyes ‘thirst’ for colored expanses; if I see a colored shape, my thirst is satisfied. My visual acquaintance with the colored expanse is the event that satisfies the perceptual drive. When I have the colored expanse in proper view, I take pleasure in my perceiving and, derivatively, in the object perceived. We have then our answer: acquaintance is not knowledge of, but enjoyment of things:
If we insist in using a verb which takes ‘content’ as its object and the ‘ego’, or ‘mind’ as its subject, the word ‘enjoying’ presents itself. It is the nearest equivalent to the German ‘erleben’, but has certain disadvantages; we shall have to say, for instance, that the mind ‘enjoys pain’. But as we know, there is no way of speaking correctly here, we must be satisfied to banish the word ‘knowledge’ from these phrases. (Schlick Reference Schlick, Friedl and Rutte1932: 219; marked as addition to the manuscript)
For Schlick, ‘S is acquainted with O’ and, in turn, ‘S is directly aware of O’ should be glossed as ‘S enjoys O’. Acquaintance with O is taking sensory pleasure in O because our perceiving O satisfies our drive for the free exercise of our senses.
Taking sensory pleasure in something is a form of evaluation. Hence, acquaintance is apprehending a value by taking sensory pleasure:
Whoever comes close to things participates in their ways and works is engaged in living, not in knowing; to him, things display their value aspect [Antlitz ihres Wertes], not their nature. (Schlick Reference Schlick1925: 80; in part my translation)
6. From the Aim of Cognition to its Metaphysics
We have now seen why Schlick endorses (P2) of the Argument from the Metaphysics of Cognition. Let us then tackle the first premise of this argument. Why does knowledge require comparison of something with something else?
In order to answer this question we must find a necessary feature of knowledge that implies that every instance of knowledge is a comparison (in a suitably broad sense). What can this feature be? Schlick takes Avenarius to have given the core of the correct answer:
This element, moreover, has already been previously recognized as an essential one—by Avenarius, among others, who says, that in apprehending, the mind ‘reduces the unapprehended to something that, as apprehended, is already part of the mental stock’. (Schlick Reference Schlick1913: 145 [478])
Knowledge acquisition is distinguished from other mental processes by its effect: the reduction of something new to something familiar. Schlick takes this initial characterization to be in need of refinement, but the refinements are not of interest for our purposes.
Let us get clearer about the common element in knowledge by looking at Avenarius's book whose title Philosophy as Thinking of the World in accordance with the Principle of Least Effort: Prolegomena to a Critique of Pure Experience (Philosophy als Denken der Welt gemäss des kleinsten Kraftmasses: Prolegomena zu einer Kritik der reinen Erfahrung [1876]) gives us a very good sense of what it is about. According to Avenarius, the soul is something that has the power and the function to enable a body to survive by adequately responding to what is going on in its environment. Since its power is finite, the body tries to expend as little as possible of it on this task. One way to minimize expenditure is forming habits of thinking: brute mechanisms to respond to the same inputs in the same way without additional cognition (see Avenarius Reference Avenarius1876: 8).
What happens if there is a new perception or presentation for which the soul has no habitual response yet?
The soul received the given presentation, but transforms what is unfamiliar in it into something familiar. In other words, it reduces by means of association the new to the old, the uncommon to the common, the unknown to the known, the ungrasped to what as something grasped is already owned by us intellectually. Now thinking a new presentation no longer requires significant expenditure of power because the soul thinks it by means of the familiar presentations. (Avenarius Reference Avenarius1876: 9–10, my translation)
The function of the activities of the soul is to transform the unfamiliar into the familiar in order to minimize effort (see Avenarius Reference Avenarius1876: 14). Avenarius calls this activity ‘begreifen’.
Schlick follows Avenarius in holding that acquiring knowledge is ‘begreifen’ and ‘begreifen’ is reducing something new (unknown) to something old (known). The ‘reduction’ is effected by comparing a new object n to already known objects (Os) and exercising the habits one has with respect to Os also with respect to n. To use Schlick's stork example, if I ‘compare’ this thing to a stork, I will expect that it makes the sounds storks have made, and so on. In short: Comparing puts me in a position to apply my dispositions for responding to (as well as my knowledge of) the behavior of known objects to yet new objects while saving me from having to make new experiences. In this way comparing enables prediction, and prediction enables successful action that, in turn, serves our drive to avoid pain and seek out pleasure. Schlick quotes Comte's slogan, ‘savoir pour prévoir et prévoir pour pouvoir’ (see quote below). We seek knowledge in order to have foreknowledge of events, and we seek foreknowledge of events in order to control them, and we want to control future events to survive and thrive.
Schlick (Reference Schlick1925: 95ff. [88ff.] identifies the future action-enabling role of knowledge as its primary aim. However, activities undertaken for one aim can become intrinsically valuable. Cooking was initially done to produce sustenance. Now many people cook for the sake of cooking. The same goes for knowledge acquisition (see also Schlick Reference Schlick, Friedl and Rutte1932: 212): it can become an end in itself.
Schlick goes an important step further than Avenarius by trying to shed light (Reference Schlick1925: 94 [87]) on the nature of knowledge via its practical aim. Our concept of knowledge is that of a state that enables successful future-oriented action. In ‘Form and Content’ this idea comes to the fore:
In every case the practical aim of knowledge is prediction, and we have good reasons to regard as the defining characteristics of knowledge those properties of it which make prediction possible. ‘Savoir pour prevoir’. (Schlick Reference Schlick, Friedl and Rutte1932: 211)
The properties of a mental state that enable it to have predictive power are those that make it knowledge. A necessary condition for a mental act to have predictive power is that it is awareness of a resemblance between an object and objects one has already encountered. Hence, only a mental state that has two terms can be knowledge. A mental state that has only one term is by definition not knowledge! It lacks the properties that make prediction possible. We have now an argument for the first premise of the Argument from the Metaphysics of Cognition:
The Aim of Knowledge Argument
(P1) The practical aim of knowledge is prediction.
(P2) The defining marks of the concept of knowledge are those properties knowledge must have to make prediction possible.
(P3) In order to predict how an object will behave, one must compare it to objects one already knows.
Therefore: (C) It is a defining mark of knowledge to be a mental relation between at least two terms.
By this argument a two-place relation to an object like Russell's acquaintance cannot be knowledge.
7. Knowledge, Foreknowledge, and Factivity
Is Schlick's case against intuitive knowledge convincing? In order to answer this question let us have a look at the Aim of Knowledge Argument. According to this argument, a mental state is knowledge only if its purpose is to enable predictions. The constitutive features of knowledge are those features of a mental state that make prediction possible.
Now it is plausible that knowledge often enables us to make predictions. For example, if I know that x is a dog, I am in a position to make a number of predictions about x’s future behavior. Such knowledge has practical value. But Schlick's teleological view of knowledge goes further and takes the concept of knowledge to be that of a state whose features enable predictions. However, there are paradigm examples of knowledge that have no obvious predictive power or purpose. For example, I know that Hadrian's wall was begun in AD 122. I also know that I exist and that I had two slices of toast for breakfast. What can I predict on the basis of such knowledge? There seems to be no independent reason to say that such knowledge is botched or not ‘real’ knowledge. Propositional knowledge can be of any facts whatsoever.
Schlick suggests that knowledge that does not yield predictions is not useful: ‘The immediate presence of certain contents is doubtless an enjoyable thing: but knowledge creates order, and that is a useful thing’ (Schlick Reference Schlick, Mulder, Kox, Hegselmann, Mulder, Kox, Hegselmann and Heath1933–34: 74 [104]). But imagine that you were never directly aware of red. Without further aids you will not be able to make predictions about objects that depend on recognizing that they are red. Insofar as acquaintance with features is involved in tracking objects, this acquaintance is useful and should not be discarded. Similarly, my knowledge that Hadrian's wall was begun in AD 122 may come in handy in many circumstances without enabling prediction. I can, for example, explain past events by appealing to this knowledge.
There is a further, more explanatory reason why Schlick's aim of knowledge argument is unsatisfactory. Some cases of knowledge yield predictions, but so do true beliefs. If I believe truly, but do not know, that the object coming toward me is a Rottweiler, I am right to expect that it will bite me. One may even truly guess future behavior. Schlick's teleological theory of knowledge cannot distinguish between foreknowledge and other mental states that enable prediction.
Now, the view that it is an essential feature of knowledge to be a factive state distinguishes knowledge from true belief and similar states independently of any purpose. If my belief that p is true, it is the case that p. But a true belief is not a kind of mental state, it is a mental state with an additional property, truth, and this property guarantees that it is the case that p. In contrast, the failure of attempts to analyze knowledge makes plausible that it is a conceptually and metaphysically simple factive state. Knowledge has practical value for us, and we rightly rely on it when making predictions because it is a state ‘whose essence includes a matching between mind and world’ (Williamson Reference Williamson2000, 40). If we know that p, we are safely right about p, and therefore our predictions are better than those based on mere true belief. The factivity view of knowledge captures cases of knowledge that do not underwrite predictions. If I know that Hadrian's wall was begun in AD 122, it is a fact that Hadrian's wall was begun in AD 122. I am in this state, for instance, in virtue of having learned by testimony that Hadrian's wall was begun in AD 122.
The factivity view also unifies knowledge of things and propositional knowledge (see section 2). Both are mental states one cannot be in if what the states are directed at does not exist. In the case of knowledge of things, being in the state guarantees the existence of an object; in the case of propositional knowledge that p, being in the state guarantees that the fact that p exists.
In sum: we have a reason to resist Schlick's claim that knowledge can be defined in terms of features that enable it to yield predictions. Knowledge that enables us to predict future behavior is of special importance for us. But a mental state is knowledge independently of whether it enables us to make predictions; it is knowledge if its essence includes a matching between mind and world.
Schlick may of course integrate the ‘matching feature’ into his account of knowledge and use it to explain the predictive success of knowledge, but if he does so he no longer has a reason to exclude acquaintance from knowledge. When I am directly aware of magenta by seeing, I know magenta because my mental state essentially matches the world; this knowledge is objective in the sense discussed in section 2.
With this in mind, let us look at a case of acquaintance. According to Schlick, when I immerse myself in an object, I ignore its resemblance to objects I know, and I maintain that state because I take pleasure in it and its object. This may be right and even phenomenologically plausible. However, it does not rule out that acquaintance is knowledge of a thing. If I am aware of an object, say, a shade of color, and I manage to focus on it and nothing else, I am still in a factive state whose essence it is to match the world.
Schlick's responds to someone who claims that acquaintance with a red patch is knowledge of the nature of the color seen as follows:
When I give myself fully to an intuitive content of my consciousness, say, a red patch I see before me, or when in behaving I submerge myself fully in the feeling of activity, I experience through intuition the red or the activity. But have I really come to know the essence of the red or of the activity? Not at all. If I had arranged the red in some order through a comparison with other colours and thereby correctly designated its shade, if I had analysed the feeling of activity psychologically and discovered in it, for example, sensations of tensions, feelings of pleasure, and so on—then I could assert that I had come to know to a certain degree the nature of the red I experienced or the feeling of activity. In intuition objects are only given, not grasped [begriffen]. (Schlick Reference Schlick1925: 82–83 [77]; I have changed the translation of ‘begreifen’ from ‘understanding’ to ‘grasping’)
Imagine that you have managed to immerse yourself in a shade of magenta. What do you gain if you manage to compare or associate with previous colors you have seen? Schlick wants to say that only through ‘compare and contrast’ with other colors do we come to know the essence of the color shade. This seems highly implausible. The fact that the color is like other colors, that it is designated by the English word ‘magenta’ or falls under a color concept has nothing to do with its essence. It may be metaphysically necessary that the color falls under some concept or other. But it is not essential that it falls under a concept and so on. We do not need to mention concepts to say what the color is.
Russell seems to have this in mind when he writes:
The particular shade of colour that I am seeing … may have many things to be said about it—I may say that it is brown, that it is rather dark, and so on. But such statements, though they make me know truths about the colour, do not make me know the colour itself better than I did before: so far as concerns knowledge of colour itself, as opposed knowledge of truths about it, I know the colour perfectly and completely when I see it and no further knowledge of itself is even theoretically possible. (Russell Reference Russell1912: 25)
Comparing and episodic recall are neither necessary for propositional knowledge nor for knowledge of objects and their essences. Yet, they have a role to play in a different respect. When I compare a color shade with other shades I already know, I put myself in the position to come to know relational truths about it and to pay attention to the color shade itself. In the case of the feeling of activity, comparing it with other feelings enables me to discern and attend to features of it. Thus, we can increase our knowledge of truths about the objects of acquaintance. This is compatible with the view that we already knew these objects before we knew truths about them. We can know truths about them because we can pay attention to parts or features of things we know.
Accordingly, immersion in an object does not prevent knowledge of it. I can be fully immersed in an object and take sensory pleasure in it as well as know it. But if I am fully immersed in an object, I will not be driven to compare and contrast it with others. I will not be driven to apply what I already know about the object and analyze it and its constituents. This goes against the very thought of economy philosophy that all knowledge is the result of the drive to economize. But we have already seen that there are independent grounds to doubt that knowledge is conceptually tied to prediction and, more generally, to economizing.
8. Conclusion
We have now seen why Schlick thinks there is a dichotomy between intuition and knowledge and why it is cashed out in part by the ‘one term’ versus ‘two term’ terminology. The dichotomy between intuition and knowledge rests on a dichotomy between two basic and incompatible drives: the drive for prediction that manifests our will to live and the drive for pleasure. The purpose of knowledge is to satisfy the first drive, and the purpose of acquaintance is to satisfy the second drive. If we are acquainted with something, we have managed to overcome the first drive and the result no longer serves the purpose of knowledge. Schlick concluded that acquaintance cannot be knowledge.
The problem with Schlick's argument lies in his conception of knowledge. We cannot merely characterize knowledge as a mental state that serves the will to live by enabling predictions. Schlick lacks a theory about the distinction between knowledge and other attitudes that can make predictions rational. If we add an independently plausible conception of knowledge, his arguments to the effect that acquaintance cannot be knowledge fail.
Consequently, Schlick does not give us a good reason to distinguish between propositional and conceptual knowledge on the one hand and nonpropositional and nonconceptual acquaintance, something entirely nonepistemic, on the other hand. We are not forced to think that we either participate in life and experience values or know propositions about relations between properties. Instead, there is a distinction between two kinds of knowledge: irreducible knowledge of objects and propositional knowledge. How these kinds of knowledge are related to each other and what the nature and value of knowledge of objects are—these are pressing and important questions. They have largely gone unrecognized and unexplored by philosophers who accepted the dichotomy Schlick argued for. If we have no reason to follow Schlick's lead in this point, we can start to pursue the right questions.
