Locke’s Theory of Ideas and the Myth of the Given Martin Lenz (Groningen) Imagine that you look around and notice a green apple on the table in front of you. Given this perception, it seems quite natural to assume that you also know that there is a green apple on the table. The move from a perceptual state to a knowledge claim is quite effortless. We regularly justify such claims by appealing to our sense perception: «I know that the apple is on the table, because I’ve just seen it there.» But what exactly is it that we appeal to? So-called foundationalists might say that our sense experience forms a privileged basis of knowledge, and then give a reason as to why sense experience is a particularly secure ground or perhaps even the only ground of true knowledge claims. One such argument is that sensations or sense-data are naturally caused and thus immune to error. So-called coherentists might object that this is mistaken; for even if we take ourselves to be appealing to sense experience, we do in fact rely on something that is already conceptual and not mere sense experience. Even if we assume that we rely on what we see, what we in fact ‘see’ is already conceptualized and propositional. It is not a raw sense-datum that we notice; rather we already ‘see’ something as a green apple. Accordingly, our appeal to (conceptualized) sensations is not immune to error and has thus a normative dimension. Now, who is right? Before addressing this question we should first take a step back and see what the discussion is about. Prima facie, the discussion between foundationalists and coherentists resembles the debate between what we take to be classical empiricists and rationalists. Yet, there is an important difference. While the latter debate is – inter alia – about whether knowledge is rooted in sense experience or in (innate) reason, the younger debate between foundationalists and coherentists has shifted in focus. The objection raised against the foundationalists does not necessarily include the denial of an empirical foundation of knowledge; rather it brings up the question whether the foundationalists’ seamless move from the natural (sensations) to the normative (concepts) is possible. In pointing out that knowledge claims cannot but rely on conceptual activity, coherentists diagnose that foundationalists fall prey to what Wilfrid Sellars called the «Myth of the Given», i.e. the assumption that knowledge can be justified by appealing to raw sensations, produced in the knower by merely causal interactions with the world1. 1 See for an opinionated but succinct discussion J. McDowell, Mind and World, Harvard, Cambridge/Mass. 1996, pp. xi-xxiv. 1 So, asking who is right can mean various things. It can simply mean to ask whether the foundationalists or the coherentists defemd a good theory. But one can also remain agnostic about that question and simply ask whether the coherentists’ objection is right, i.e. whether foundationalism falls prey to the Myth of the Given. It is the latter question I want to pursue in this paper. I will do so by investigating whether the perhaps most famous foundationalist, namely John Locke, has fallen prey to that Myth. What is philosophically at stake here? In the wake of Sellars, it is now common to distinguish between the causal order of nature and the normative order of the conceptual. Although it is difficult to draw a clear line of demarcation between the «space of nature» and the «space of reasons», it seems to govern a host of significant distinctions, such as those between causes and reasons, facts and values, determination and freedom, and sensation and thought or sensation and knowledge. Thus, «falling prey to the Myth» means to commit a serious mistake, namely in ignoring that there is a dividing line between the natural and the normative. Although one might question how exactly that line is to be drawn or even question whether there is a real separation between these realms, it is crucial to see whether Locke’s theory of ideas can accommodate the pertinent differences. Locke’s theory of ideas is indeed a common example of a position falling prey to the Myth. As is well known, Locke’s Essay is mainly an epistemological work and attempts to discuss the origin, certainty and boundaries of human knowledge. The point that Locke wishes to establish is that all our knowledge is founded on experience. ‘Idea’ is a technical term that denotes the kind of cognitive vehicle that mediates all our cognitive activities. So if we were to give a pertinent technical description of what is involved in noticing an apple we would say that we have a complex idea of an apple. According to Locke, such a complex idea can be analysed into simple ideas that, in turn, are caused by external properties and thus might be taken as a secure foundation of knowledge. In what follows, I will argue that Locke, although he clearly has foundationalist ambitions, does not fall prey to the Myth. To do so, I shall begin by characterising the Myth. Secondly, I will show how the Sellarsian objections can be launched against Locke’s theory of ideas. Thirdly, I will present my interpretation of Locke’s take on ideas and show how they fare in relation to the crucial features involved in the discussion of the Myth2. By way of conclusion, I will discuss whether Locke’s way of avoiding the Myth limits his foundationalist approach. 2 My interpretation heavily draws on M. Lenz, Lockes Sprachkonzeption, De Gruyter, Berlin – New York 2010, pp. 215-349, see esp pp. 254-263. I wish to express my gratitude to Markus Wild for his valuable comments to an earlier version of this paper. 2 1. The Myth of the Given As Wilfrid Sellars himself points out, there are of course several forms of the so-called Myth of the Given3. In one of its crucial varieties, it is the idea that empirical knowledge can be justified with recourse to immediately perceptible sense impressions. So the Myth might be said to consist in the combination of the following two theses: 1. The mind can passively receive sense impressions that are not influenced by any conceptual or inferential activity. 2. Such sense impressions can justify empirical knowledge. If we accept the combination of these theses, we can say that someone having a sensation of a green patch has thereby knowledge of there being a green patch. Accordingly, we might assume that this myth is a particular problem for epistemologists. In fact, however, the problem runs deeper. What is at issue is not only the inference of knowledge from given sensations but, more generally, the inference of conceptual content from the awareness of something. This means that having a sensation of a green patch can entail thinking of this as a green patch, without presupposing prior possession of the concept of green. Sellars puts this as follows: «If a person is directly aware of an item which has the categorial status C, then the person is aware of it as having categorical status C. This principle is perhaps the most basic form of what I have castigated as “The Myth of the Given”.»4 In this basic form, the Myth of the Given or Givenism is a problem not only for epistemology but quite generally for any theory of the mind that allows for inferences from naturally caused percepts to concepts5. What discredits this idea as a myth is the confusion of the realm of the natural and causal, on the one hand, with the realm of the conceptual and normative, on the other. While the first thesis is supposedly a claim about the natural realm, the second thesis is a claim about the normative realm. In contrast to the merely causal processes of receiving sense impressions or data, justifications and other kinds of conceptual activity are normative in that they can be correct or incorrect6. In other words: mere information does not constitute knowledge. 3 4 5 6 See W. Sellars, The Lever of Archimedes, in K. Scharp / R. Brandom (eds), «In the Space of Reasons», Harvard University Press, Cambridge/Mass. 2007, pp. 229-257. W. Sellars, The Lever of Archimedes cit., p. 236 sq. If we construe the critique against givenism in this broader sense, then the charge cannot solely be countered on epistemological grounds, as C. Panaccio suggests by ascribing an epistemological externalism to Ockham in an otherwise compelling interpretation. See C. Panaccio, Ockham: Intuition and Knowledge, in B. Held / L. Osbeck (eds), Rational Intuition, Cambridge University Press, Cambridge (forthcoming). I would like to thank Claude Panaccio for letting me have a preliminary version of his paper. See J. O’Shea, Wilfrid Sellars: Naturalism with a Normative Turn, Polity Press. Cambridge, pp. 106-118. 3 But what is so attractive about this myth? Why might so many philosophers have adhered to it? Perhaps the most intriguing feature is the idea of immediacy. Something affects my sense organs and causes an idea, say an idea of something green. If I have this idea, then my knowledge that there is something green is immediately justified. This means that I do not need any further reflections or inferences, since this piece of knowledge does not depend on any other ideas or sentences; all it depends on is this given idea. A second advantage is the veracity of such a justification; it is true independently of any conceptual norms. For if knowledge is justified by causal processes, there is literally nothing that can go wrong, because representations whose content is due merely to causal processes can only be correct. By definition, such an idea is true of whatever it is that causes it. Let me illustrate this with an example. If you receive sense impressions from a green apple – say impressions of a certain green, roundness, a certain texture, smell and so on – then these impressions can serve at the same time as a justification of your having a true belief, say the belief there is an apple or the belief that the apple is such an such. Thus, givenness would imply that the mere causal contact with something would be sufficient for grasping the source of these impressions as that what it is, namely as an apple. The tenet that there are certain kinds of cognition and knowledge that are veridical independently of any further knowledge has been around for a very long time. Now, the central objection against this Myth runs as follows. Since non-conceptual material or sense-data are not epistemic but produced by a merely causal contact between the cognizer and the world, they are not available to the cognizer in the way that concepts or sentences are available. I can think ‘in concepts’ or ‘in sentences’, one might say, but not ‘in sense-data’. Thus, they cannot serve as grounds of justification or claims of knowledge. Conceptual justifications can only be founded on something conceptual, not on something non-conceptual. Someone who appeals to sensible ideas or sense-data for epistemic justification introduces material from the causal order into the normative order of the conceptual, as it were. Another way of putting this objection would be to say that an idea, let’s take the idea of green again, plays two roles at once: on the one hand, it appears in a causal explanation, namely if it is taken as something that is caused by an external property; on the other hand, it enters into the thought or sentence that is being justified, namely if it is taken as the idea or concept of green. But one idea cannot play such a dual role, and to claim otherwise is to work on the basis of an equivocation7. 7 There are of course many different lines of argument against Givenism, but I consider these points crucial for the purposes of the present paper. See for helpful reconstructions: W. deVries / T. Triplett, Knowledge, Mind, and the Given, Hackett, Indianapolis 2000, pp. xv-xlvi. 4 In the Cartesian tradition, Locke is generally taken as a prime example of someone who has fallen prey to the Myth8. The charge is that he portrayed simple ideas as immediate, passively received sense-data and at the same time as components of the complex ideas that are available to us. To be sure, this interpretation does not come out of the blue, for Locke writes: «These simple Ideas, the Materials of all our Knowledge, are suggested and furnished to the Mind, only by those two ways above mentioned, viz. Sensation and Reflection. When the Understanding is once stored with these simple Ideas, it has the Power to repeat, compare, and unite them even to an almost infinite Variety, and so can make at Pleasure new complex Ideas.»9 On a straightforward reading of these and other passages, Locke seems to suggest that the understanding passively receives sense impressions, is thus stored with simple ideas from which the understanding can form complex ideas. Accordingly, there is a move from simple ideas qua sense-data to simple ideas qua concepts. So, given a certain amount of simple ideas (say of round, green and firm) my mind can form the complex idea of a green apple from this material. Thus, the causal space of nature is confused with the epistemically available space of the conceptual. Accordingly, the epistemic foundation of knowledge would rest on an equivocation in the notion of simple ideas or, more generally, in the notion of experience, since simple ideas would play a central role in both spaces10. In the contemporary discussions on the Myth of the Given, there are commonly two opposing ways of getting around this equivocation. Naturalists claim that all conceptual activity that purportedly takes place in the space of reasons can in fact be traced back to causal processes. By contrast, coherentists claim that all conceptual activity takes place in the space of reasons. Taken to their extremes, both ways are problematic. The first way is problematic in that a complete naturalization of the mental cannot accommodate our common intuitions 8 9 10 See, for instance, S. Blackburn, Truth: A Guide for the Perplexed, Allan Lane, London 2005, pp. 140-144 and F. Kambartel, Erfahrung und Struktur: Bausteine zu einer Kritik des Empirismus und Formalismus, Suhrkamp, Frankfurt/Main 1968, pp. 15-49. Although Cartesians and certain subsequent philosophers are considered prime targets of the critique, I am not convinced that medieval authors are immune to the charges simply by not endorsing a mind-body dualism, as Peter King claims. See P. King, Medieval Intentionality and Pseudo-Intentionality, «Quaestio», 10 (2010), p. 43. For, as we shall see, the critique also targets positions that rely on the immediacy, non-propositionality, abstractability and the causal nature of sense impressions. See J. Locke, An Essay Concerning Human Understanding, ed. Peter H. Nidditch, Clarendon Press, Oxford 1975, II, ii, 2, p. 119. See L. Krüger, Der Begriff des Empirismus. Erkenntnistheoretische Studien am Beispiel John Lockes, De Gruyter, Berlin – New York 1973, pp. 19-39, for an insightful discussion of the exegetical issues involved in the interpretation of the notion of simple ideas. 5 about judgments, consciousness or freedom. The externalist suggestion, for instance, according to which justification is sufficiently accounted for if our knowledge is based on reliable processes and thus does not have to rely on the awareness of the knower itself, does not sit easily with the idea that, when ascribing knowledge to persons, we would equally want to ascribe to them justifying reasons available to themselves. The second way is problematic in that a complete coherentism threatens to leave us without any empirical foundation of epistemic justification11. For if my thoughts are not to be justified with recourse to sense experience but only with regard to other thoughts or sentences, then the world’s friction seems to be lost. To be sure, the host of problems raised by this opposition between naturalism and coherentism admits of various ‘third ways’ whose upshot often consists in connecting the realms of nature and norms without falling prey to a straightforward equivocation incurred by appealing to «the given». In what follows, however, I do not wish to discuss the third ways as taken by Sellars, McDowell and others. Rather, I hope to show how Locke’s theory of ideas fares with regard to the Myth of the Given. 2. The Givenness of Ideas and Sellars’ Critique Before looking more closely at Locke’s position, we need to address a methodological problem. For already at this point one might object that the question whether Locke fell prey to the Myth of the Given is owing to an anachronistic reading. The discussion of Givenism seems to be a characteristic of 20th century pragmatism rather than part of the epistemological projects of the 17th century. Prima facie, then, this kind of objection seems justified. As for instance Rorty has famously argued, Locke and other epistemological foundationalists in the wake of Descartes considered the cognitive faculties mainly with regard to question of how much they achieve to «mirror» nature and failed to acknowledge the normative aspects inherent in this model of cognition. But this general picture of early modern epistemology does no longer stand up to critical scrutiny. Indeed many early modern philosophers can be seen as attempting a carefully balanced consideration of the relation between the natural and the normative12. Moreover, it is evident that Locke himself rejected certain kinds of myths of the given. Firstly, he famously challenged the tenet that we are endowed with innate principles and ideas that are epistemically available to us merely on account of our mind’s possessing them. 11 12 See for a general discussion J. McDowell, Mind and World cit., pp. xi-xxiv. On the internalist-externalist debate, see the introduction in R. Schantz (ed.), The Externalist Challenge, De Gruyter, Berlin – New York 2004. See for instance the introduction to M. Lenz / A. Waldow (eds), Contemporary Perspectives in Early Modern Philosophy: Nature and Norms in Thought, Springer, Dordrecht – London 2013. 6 Indeed, one of Locke’s points is that, even if some principles were innate, their mere presence would not explain anything, since they still would have to recognized or activated and used. So Locke does by no means mix up the possession of ideas with their epistemic availability and use. These same is of course true for the material provided by the senses; neither innate nor acquired material does as such amount to any epistemic activity; rather senses and reason always have to interact. In other words: the merely causal reception of material does not constitute knowledge or other kinds of normative conceptual activity. Already in his rarely studied Questions concerning the Law of Nature Locke points out explicitly that reason and senses mutually depend on each other13. Secondly, he refuted the assumption that we possess substance concepts independently of our cognitive development14. In this case, the problem is that, although we obviously use substance concepts of which we predicate certain properties, we cognize the supposed substances only by means of the properties: so whence do we get the concept of substance if it is not innate? Locke does see the problem but chooses a different route of explanation. It is our mind’s repeated encounter with patterns of ideas of properties that makes us subconsciously ascribe these patterns to one thing, namely a substance. So, again, it is not any innate or given concept but the development of cognitive habits in response to encounters with perceptual patterns that Locke invokes to explain this ability of the mind. In both cases, then, Locke thinks that the mere givenness of concepts would be explanatorily idle. Rather, it is the interaction of our cognitive apparatus with the material that leads to what we would call conceptual activity. And although these examples are still fiercely discussed, it is certain that they display the kind of doubt that also motivates later worries about assumptions of concepts or principles just given to us. 13 14 See J. Locke, Questions concerning the Law of Nature, ed. R. Horwitz / J. Strauss Clay / D. Clay, Cornell University Press, Ithaca 1990, pp. 152-154: «Quae dum mutuas sibi invicem tradunt operas, dum sensus rerum particularium sensibilium ideas rationi administrat, et suggerit discursus materiam, ratio e contra, sensum dirigit et ab eo haustas rerum imagines inter se componit, alias inde format, novas deducit, nihil tam obscurum est, tam reconditum, tam ab omni senu remotum, quod his adjutus facultatibus, cogitando et ratiocinando assequi non possit omnium capax animus. quod si alterutram tollas, altera certe frustra est …» Locke’s formulation does indeed bear some resemblance to the Kantian dictum «Thoughts without intuitions are empty, intuitions without concepts are blind.» See also J. Locke, Essay cit., I, passim. See J. Locke, Essay cit., II, xxiii, 14, p. 305: «These Ideas of Substances, though they are commonly called simple apprehensions, and the Names of them simple Terms; yet in effect, are complex and compounded.» See also Essay II, xxiii, 1: 295: «The Mind being, as I have declared, furnished with a great number of the simple Ideas, conveyed in by the senses, as they are found in exteriour things, or by Reflection on its own Operations, takes notice also, that a certain number of these simple Ideas go constantly together; which being presumed to belong to one thing, and words being suited to common apprehensions, and made use of for quick dispatch, are called so united in one subject, by one name; which by inadvertency we are apt afterward to talk of and consider as one simple Idea, which indeed is a complication of many Ideas together; Because, as I have said, not imagining how these simple Ideas can subsist by themselves, we accustom our selves, to suppose some Substratum, wherein they do subsist, and from which they do result, which there7 But despite Locke’s explicit attacks against given ideas in the sense of innatism, the question remains whether Locke does not still fall prey to the Myth of the Given by tacitly taking simple ideas at once as products of causal processes and as elements of epistemic justifications in normative contexts. At first, it seems as if Locke’s theory of ideas does indeed fail on this account. In fact, the claim that the passive reception of certain simple ideas – let’s say the idea of green – in the causal contact with pertinent properties produces general concepts seems to be all over the place. This suggests a typical bottom-up model of cognition: in a first step of the cognitive process, our mind’s particular ideas causally covary with certain properties in the world. In a second step, we obviously can abstract a general idea or concept of green from this given idea. This idea is general in that it matches several instances of green and can be associated with the word ‘green’. However, the abstract idea of green is no longer a mere sense-datum or particular idea; rather our mind can use it as a sign of its cause, i.e. of the occurrence of a certain property. Locke writes: «This is called ABSTRACTION, whereby Ideas taken from particular Beings become general Representatives of all of the same kind; and their Names general Names, applicable to whatever exists conformable to such abstract Ideas. … Thus the same Colour being observed to day in Chalk or Snow, which the Mind yesterday received from Milk, it considers that Appearance alone, makes it a representative of all of that kind; and having given it the name Whiteness, it by that sound signifies the same quality wheresoever to be imagin’d or met with; 15 and thus Universals, whether Ideas or Terms, are made.» According to this account, it seems that the first step of cognition consists in a causal production of particular ideas of white that are abstracted for cognitive use in a second step. The critical passage from the causal space of nature to the normative space of reasons occurs in the second step. Since it is here that we supposedly attain a concept whose presence is taken to allow for correct or incorrect judgments about the occurrence of white properties. Although this consideration regarding the reception and abstraction of simple ideas indicates no more than a sketch of Locke’s theory it allows us to highlight four aspects that seem crucial to any epistemological foundationalism prone to falling prey to the Myth of the Given: 1. the aspect of immediacy, i.e. the immediate givenness of particular ideas of properties; 15 fore we call Substance.» For a succinct interpretation, see L. Newman, Locke on the Idea of Substratum, «Pacific Philosophical Quarterly», 81 (2000), pp. 291-324. See J. Locke, Essay cit., II, xi, 9: p. 159. 8 2. the negligence of propositionality, i.e. the assumption that we could epistemically relate to contents such as ‘green’ without their being embedded in a propositional form; 3. a circular abstractionism, i.e. the tenet that we can abstract the pertinent concepts from sense impressions without already possessing a conceptual access; 4. the underestimation of normativity, i.e. the tenet that such an abstraction does not only result in concepts but in adequate concepts without invoking any social practice that governs notions of adequacy and correctness. In the following section, I will discuss these aspects one by one and see whether Locke’s theory takes them into account. I shall argue that Locke’s answer to the questions related to the Myth of the Given does not rely on any equivocation in his theory of ideas. Rather, it will become clear that his solution is founded on what I would like to call a dual perspective of explanation. Before moving on, however, we should look more closely at the objection against Locke. The example of perceiving a colour such as green or white might sound quite harmless, but if the sketched reconstruction is correct, the very foundations of Locke’s theory fall prey to the Myth. According to Sellars, already the first step of the cognitive process, i.e. the reception of the sense-datum as a particular idea of green, would presuppose conceptual abilities, while he takes Locke to assume the contrary: «… Locke mistakenly concluded that concepts and categories could be abstracted from sheer sensibility. Kant infers that the framework of basic concept and categories is innate.»16 Accordingly, already the mere reception would require the very concept of green that is supposed to be abstracted in the second step. In his famous Empiricism and the Philosophy of Mind, Sellars puts this point very succinctly: «For now we recognize that instead of coming to have a concept of something because we have noticed that sort of thing, to have the ability to notice a sort of thing is already to have the concept of that sort of thing, and cannot account for it.»17 So if I want to move from the cognition of a particular sense-datum to an abstract idea, I already have to have a conceptual grasp of that sense-datum. Locke’s theory of ideas, however, 16 17 W. Sellars, Some remarks on Kant’s theory of experience, in K. Scharp / R. Brandom (eds), «In the Space of Reasons», Harvard University Press, Cambridge/Mass. 2007, p. 449. W. Sellars, Empriricism and the Philosophy of Mind (1956/1963), reprinted in W. de Vries / T. Triplett, Knowledge, Mind and the Given: Reading Wilfrid Sellars’s “Empiricism and the Philosophy of Mind”, Hackett. Indianapolis – Cambridge 2000, p. 256. 9 does not seem to account for this prior conceptual grasp, if it merely appeals to causal covariance18. 3. Locke’s Dual Perspective This critique would be fatal, if it were indeed applicable to Locke’s position. But this is not the case. If we look carefully at Locke’s account we will see that the two steps sketched above are not to be taken as steps of the cognitive process but as two different explanatory perspectives, namely a natural historical and an epistemic perspective. This point, though crucial for any interpretation of Locke’s theory of ideas, is often overlooked. In his Essay concerning Human Understanding, the notion of simple ideas is explicitly introduced twice. Locke begins by accounting for their causal route into the mind according to his «Historical, plain Method» and then re-introduces them with regard to their relation to other ideas: «Though in the foregoing part, I have often mentioned simple Ideas, which are truly the Materials of all our knowledge; yet having treated them there, rather in the way that they come into the Mind, than as distinguished from others more compounded, it will not be, perhaps amiss to take a view of some of them again under this Consideration …»19 Thus, the first step, or rather perspective, provides an answer to the question of the origin of ideas. Locke’s point is that ideas are not innate but that they arise from the causal contact with external things or properties. The second perspective provides an answer to the question of the epistemic content of ideas, that is: how are the ideas related to one another in our minds, and what kind of epistemic content do they have for us within conscious episodes of thinking? It is with regard to the second question that Locke notes that ideas have content inasmuch as they are general signs of a property of something. So while the first perspective accounts for the causal history of ideas, the second perspective reveals the content of ideas as concepts of substances, modes or relations. In this sense, the ideas that were previously treated as simple 18 19 See for a critical discussion of Locke’s representationalism as a causal covariance theory S. Ferguson, Lockean Teleosemantics, «Locke Studies», 1 (2001), 105-122. See J. Locke, Essay cit., II, xiii, 1, p. 166. Locke even clearly signals the ending of the merely natural historical account. See Locke, Essay cit., II, xi, 15: 162: «And thus I have given a short, and, I think, true History of the first beginnings of Humane Knowledge; whence the Mind has its first Objects, and by what steps it makes its Progress …» Although many commentators stress the importance of the natural historical method, they mostly ignore Locke’s change of perspective. Even the most recent discussion, though acknowledging that the theory of ideas needs to be reconsidered, does not take into account Locke’s own remarks on the re-introduction of simple ideas. See A. Nelson, Philosophical Systems and Their History, in 10 are now re-considered as constituents of complex ideas that are, in turn, constituents of thoughts or mental propositions. Locke puts this point as follows: «Complex Ideas, however compounded or decompounded, though their number be infinite, and the variety endless, wherewith they fill, and entertain the Thoughts of Men; yet, I think, they may be all reduced under these three heads. 1. Modes. 2. Substances. 3. Relations. »20 If we want to explain ideas with regard to the second perspective it is not sufficient to appeal to a causal explanation. Rather, as Locke makes clear in the course of his Essay, our complex cognitive development has to be taken into account. The reason is that only the recourse to our cognitive development accounts for some crucial aspects regarding the epistemic content of our ideas: firstly, there are already habitualized mental operations (such as memory and abstraction) involved that we normally do not take any notice of. Secondly, these operations are affected by social factors that affect the determination of content21. Accordingly, the first perspective, i.e. the answer to the question of origin, is not supposed to explain the cognition of a sense-datum but its causal production. But the causal production of an idea does not entail that the idea is epistemically available to us. Neither can, according to Locke, the innatists infer the epistemic availability from the innateness of ideas. If this general take on Locke’s theory of ideas is correct, then the central exegetical mistake of Sellars and numerous other commentators consists in their taking the first step of causal production as an epistemological explanation. In fact, there is no equivocation in Locke’s notion of simple ideas. Rather, the two steps mark a careful distinction between two explanatory perspectives that acknowledge the difference between the causal origin and epistemic use of ideas and thus between the space of nature and the space of reasons. In order to understand Locke’s theory of ideas we carefully need to establish what exactly it is that is being explained: the causal origin of ideas or the epistemic content that occurs in our conscious episodes of thinking. Accordingly, we can characterize the explanation of the origin as a bottom-up perspective and the epistemic explanation as a top-down perspective. When looking at episodes of thought and perception as well as epistemic justification we 20 21 M. Laerke / J. Smith / E. Schliesser (eds), Philosophy and Its History: Aims and Methods in the Study of Early Modern Philosophy, Oxford University Press, Oxford 2013, pp. 236-257, esp. p. 250 sqq. J. Locke, Essay cit., II, xii, 3, p. 164. See for a very condensed elaboration on the subconscious workings of mental operations and the social determination of content Locke’s chapter on the association of ideas in the Essay II, xxxii. Cf. M. Lenz, Lockes Sprachkonzeption cit., pp. 263-277. 11 are always already taking the top-down perspective, as it were. Let’s now assess this dual perspective with regard to the four mentioned aspects of the Myth of the Given. Immediacy Immediacy is certainly one of the most common aspects of Givennism. The assumption seems to be that ideas are given to us in thought immediately in that we require no other information or capacities in order to know what is given. But does this description fit the way Locke conceives of the way we think? Let’s quickly remind ourselves what episodes of thoughts or «trains of ideas» (to use the Lockean term) are. According to Locke, our thoughts mostly proceed in sentences that are connected to complex ideas; perceptual episodes are equally complex. The ideas, then, are not present to as unrelated «atoms of givenness» («Gegebenheitsatome»)22, but progress as structured complexes that can be analysed accordingly, namely as ordered ideas of substances, relations and modes. Thus, you do not think «Oh, green! Oh, round!» You rather think that there is a juicy green apple or that the apple is juicy and green. Now one might want to raise the following objection against this interpretation. Although this reading is indeed suggested by the way that Locke introduces complex ideas, he often expresses contrary views. For instance he writes: «That whatsoever is so constituted in Nature, as to be able, by affecting our Senses, to cause any perception in the Mind, doth thereby produce in the Understanding a simple Idea …»23 Here, Locke clearly admits that simple ideas are immediate causal products in the mind. As I have noted above, however, one should take into account that Locke invokes two different perspectives when talking about ideas. One must not overlook the fact that the quotation just given is an unmistakable instance of the natural historical perspective on the origin of ideas. More importantly, Locke immediately contrasts this with the epistemic perspective: «These [sc. ideas] the Understanding, in its view of them, considers all as … positive Ideas, without taking notice of the Causes that produce them: which is an enquiry not belonging to the Idea, as it is in the Understanding; but to the Nature of the things existing without us.»24 22 23 24 See for instance F. Kambartel, Erfahrung und Struktur cit., pp. 26-28. Locke, Essay cit., II, viii, 1, p. 132. Locke, Essay cit., II, viii, 2, p. 132. 12 In addition to clearly reflecting the distinction of perspectives, this passage highlights the fact that the consideration of the ideas’ origin does not ensue automatically or immediately. Instead, it depends on whether we take a certain scientific attitude. Let’s illustrate this point with an example. If my mind has a bundle of ideas of properties (say those that might belong to an apple), it will take them to belong to a substance; this happens in a completely habitualized manner by means of a subconscious cognitive process25. If I make an explicit judgment and thus concentrate on certain idea of a property by thinking that the apple is green, I will view this property primarily as a property of a thing. In this mode of thinking I might be guided by all sorts of interests: I might be hungry or admire the colour or what have you. None of these interests will necessarily drive me to think of my idea of green as the sign of the property that caused it. In fact, it is mainly my cognitive development as well as emotional dispositions that govern attention in thought processes26. Thus, it is only if I take on the attitude of a natural philosopher or some other scientific attitude that I might see my idea as a cognitive vehicle that signifies its cause. The distinction of perspectives, then, clarifies at least two points. Firstly, the ideas are not immediately given to us as signs of their causes. Thus, I cannot rely on their causation as an immediate means of justifying whatever knowledge I claim to have. If we really wish to consider ideas with regard to their causal production we have to take a certain scientific stance or attitude to them. Secondly, for the description of our episodes of thinking, Locke does not set out from particular ideas but from structured bundles. As will become clear shortly, these bundles are propositionally structured. Propositionality The fact that ideas are propositionally structured is of vital importance in order to see that Locke avoids the Myth in yet another sense. While modern empiricists such as Russell or Ayer are mainly charged for their appeal to immediate sense-data, early modern authors are additionally blamed for ignoring what since Frege is taken to be the context principle, i.e. the claim that concepts or words are semantically evaluable only in the context of sentences or judgments. The basic point is that the given in the form of a simple idea or a sense-datum is not propositional and thus cannot form a premise or reason in an argument. But something 25 26 See, for instance, Locke, Essay cit., II, xxiii, 1, p. 295, on the inadvertent substance ascription as well as Essay cit, II, ix, 9-10, p. 146 sqq, where Locke suggests an explication of the unnoticed work of mental operations. See Locke, Essay cit., II, vii, 2-3, p. 128, where he states that, without interests and emotional dispositions, our ideas would drift like «unregarded shadows». 13 that cannot take the place of a premise cannot figure in epistemic claims. To illustrate the point linguistically, imagine that I should justify perceptual knowledge: asked how I know there to be a green patch, I cannot just respond by saying «green»; I rather have to say that «there is something green». The charge is, then, that by ignoring the propositionality of content, Locke and other early modern authors falsely reasoned from supposing simple items as a given27. However, this still prevalent reading disregards the fact that even before Locke philosophers engaged in intense debates on the question whether judgements are prior to concepts in the order of explanation28. As we have already seen, Locke himself clearly tackles this issue in analysing simple ideas as parts of complex structures. He states that semantic evaluation and epistemic justification are always performed in the context of implicit or explicit judgments. Locke puts this point as follows: «Though Truth and Falsehood belong, in Propriety of Speech, only to Propositions; yet Ideas are oftentimes termed true or false … Though, I think, that when Ideas themselves are termed true or false, there is still some secret or tacit Proposition, which is the Foundation of that Denomination …»29 As this passage suggests, Locke endorses what might be called an ideational context principle. Even if we do not explicitly utter propositions, we still make implicit judgments by means of what Locke calls «tacit» or «mental propositions» in accordance with the scholastic tradition. What this means becomes clear, for example, with regard to our substance ascriptions. As has been noted above, we might take ourselves to perceive certain properties such as a green colour, a roundness and firmness. However, what our mind really does in such cases is the following: it inadvertently ascribes these properties to a substance, say an apple. Thereby it performs an implicit judgment. Rather than thinking of green colour, roundness and firmness we think that there is a green, round and firm apple. Accordingly, ideas in thought episodes are 27 28 29 See W. deVries / T. Triplett, Knowledge, Mind, and the Given cit., p. xxx sqq. R. Brandom, Making It Explicit, Harvard University Press, Cambridge/Mass. 1994, p. 79, even claims that traditional semantics ignored the special role of propositions: «The pre-Kantian tradition took it for granted that the proper order of semantic explanation begins with a doctrine of concepts or terms, divided into singular and general, whose meaningfulness can be grasped independently of and prior to the meaningfulness of judgments. Appealing to this basic level of interpretation, a doctrine of judgments then explains the combination of concepts into judgments, and how the correctness of the resulting judgments depends on what is combined and how. … Kant rejects this. One of his cardinal innovations is the claim that the fundamental unit of awareness, the minimum graspable, is the judgment.» See on the structure of ideas in Locke and the late scholastics M. Lenz, Lockes Sprachkonzeption cit., pp. 277-349. See Locke, Essay cit., II, xxxii, 1, p. 384. 14 always embedded in implicit judgments or background assumptions; only in the context of such (tacit) propositions can ideas be (justifiably) related to something else and become parts of epistemic justifications. Abstractionism Even if the foregoing claims seem acceptable, the tenet that our mind engages in abstraction could be taken as an insurmountable obstacle in avoiding the Myth of the Given. I think that people who still defend the reading just criticized might wish to object something like the following: «This is all very well. For Locke, simple ideas are not immediately given but conceptually modified in some sense. Moreover, he even accepts some version of the context principle. But the decisive problem has not really been addressed at all. For ultimately Locke is committed to the view that the ideas in our thought episodes are abstracted from sense-data. According to this view, abstraction is the crucial mental operation by means of which the mind tears ideas out of the material, i.e. the ideas with regard to their origin, and thus lifts the ideas from their status of being mere material to turn them into those general ideas that are available to us in our thought episodes. Now if abstraction does indeed play that role, then it is of course vital that it renders correct results. Otherwise it would not be possible to claim that the ideas taken in their relation to the sensual properties are true ideas. The upshot is that our mind abstracts the general idea of a property (say green) because it noticed several instances of this property. So the idea is true of these properties because it was or could have been abstracted from those. But this way of reasoning is plainly circular, for the ability to notice a common feature amongst several particular things already requires a concept of that common feature. The very concept whose formation was to be explained through abstraction is required for the abstraction to take place.» Thus far my fictive opponents whose view might resemble to some extend those of Peter Geach or Wilfrid Sellars30. What can we reply to this objection? Locke’s position as well as the ongoing discussion about the so-called abstractionism is quite an intricate issue. Therefore, I would like to confine myself here to making two points. Firstly, Locke’s account of abstraction is by no means circular. It does not require any cognition or recognition of non-abstract or particular ideas in any epistemic sense. Rather, abstraction proceeds on the basis of habits in a subconscious way. Why, you might ask, is the latter point so important to stress? Well, if we imagine abstraction as a conscious process, we face two options. Either we will secretly have to presuppose the concepts that are in fact taken to be formed by abstraction and thus face the circular- 15 ity objection again; or abstraction will be a completely arbitrary process and we will have to allow for psychologically fairly absurd scenarios. Just try to imagine that you wake up one morning, get out of bed, go to the bathroom and suddenly realize that your mind has – at least today – not yet performed any abstractions and therefore not provided any general ideas or concepts by means of which you could classify the kind of experience that you are having that moment. You would be confronted, it seems, with countless impressions all of which would be particular, non-repeatable and not categorized. You would be utterly lost, unable to make sense of anything around you, thrown into an Heraclitic world of permanent flux. Where then could you possibly start to abstract? – This is of course not the kind of state that Locke thought we would find ourselves in. But it is important to see that this is the state we have to assume to be in if we took abstraction to be a conscious process in which we would meet an uncategorized manifold or mere «atoms of givenness» and then would have to consciously choose and decide what to abstract and categorize. It is this false dilemma of either circularity or absurd arbitrariness that is entailed by ignoring Locke’s dual perspective31. As I see it, discussing abstraction makes no sense if we take it to be meant with regard to the epistemic perspective. Let’s reconsider this account of abstraction under the natural historical perspective of origin. Here, we do not need to take it as a conscious or epistemic process. A helpful analogy of abstraction as a natural subconscious process is the one given by E. J. Lowe, who compared abstraction to the mechanism of a vending machine. Compare the occurrence of particular ideas of the same type with coins of a certain kind or value. Let’s now say that the vending machine accepts a range of coins of a certain value. Now, inserting coins is mainly a causal and mechanical process. The coins fit according to the way the machine is built: certain coins simply match certain slots in the machine. To make such a claim we do not need to assume any conscious operations inside the vending machine; it just happens; the machine does not need a special mechanism to recognise the coins. Nor does our mind need a special mechanism or some homunculus to recognise the pertinent stimuli of our nervous system that produce particular ideas of the same type32. To be sure, many interpretational problems remain, since Locke is not very explicit about the details of abstraction. But from the suggested point of view, we might envisage possible ways of rendering abstraction as process that is subconsciously governed by biologically deter30 31 32 See P. Geach, Mental Acts, Routledge – Kegan Paul, London ²1971, pp. 18-44 and p. 130 sq; W. deVries / T. Triplett, Knowledge, Mind, and the Given cit., pp. 53-58. See K. Saporiti, Locke und Berkeley über abstrakte Ideen, «Aufklärung», 18 (2006), pp. 113-142. F. Kambartel, Erfahrung und Struktur cit., pp. 15-49, raises both the circularity and the arbitrariness objection. See E. J. Lowe, Locke on Human Understanding, Routledge, London and New York 1995, p. 164 sq. 16 mined preferences. This way we can not only circumvent the circularity objection but also the arbitrariness objection. As we have already seen, Locke thinks that thought processes are to some extend guided by emotional dispositions. Moreover, he notes that there are some «leading qualities» such as colour and shape that play a vital role for our discriminatory abilities. Accordingly, abstraction would not have to be not arbitrary but could be spelled out as an operation adapted to our biological make-up33. The upshot with regard to avoiding the Myth is that we do not consciously guide our own abstractions; rather our conscious thought episodes always proceed in ideas that are abstract already. This leads us to the second point I wish to stress. The notion of abstraction just sketched does of course come at a price. Yet only essentialists will be worried about its amount. In contrast to Aristotelian hylomorphists’ theories of abstraction, Locke’s account of abstraction is not designed to ensure the correctness or adequacy of ideas. This means that abstraction does not figure in epistemic justifications; it simply delivers general ideas. To put it bluntly: abstraction just functions. It can be taken to work on the basis of a number of preferences that are determined by our biological make-up and perhaps fine-tuned by our cognitive development. But in contrast to what the opponents of abstractionism suppose, abstraction is not related to the correctness or incorrectness of ideas34. Normativity The question of correctness and adequacy is related to another pressing issue regarding the Myth of the Given, namely the issue of the normativity of ideas. Normativity is here not to be taken as a moral term; rather the fact that ideas and thoughts can be adequate and inadequate or true and false can be seen as a normative dimension. In this sense, thoughts can conform to or deviate from a standard with regard to which they are justifiable. Thus, one can speak of semantic and epistemic normativity of linguistic expressions and ideas. Now, the common charge against the Lockean account is the alleged presupposition that it is abstraction from the given that provides the grounds for epistemic justification. Accordingly, abstraction would have to account for the normativity of ideas. But as we have just seen, this charge is unfounded, since Locke can be taken to distinguish between the functioning of abstraction as a subconscious process and the question of the correctness of abstract general ideas taken as concepts figuring in our thought episodes. But if it is not abstraction, what does account for the normativity of ideas according Locke? As I see it, Locke’s account offers two resources to account for the normativity of 33 See J. Locke, Essay cit., III, xi, 19-20, pp. 518-519. 17 ideas: on the one hand, there is God’s design of the relation between simple ideas and the external causes; on the other hand, there are the relations between complex ideas and linguistic expressions (nominal essences) governed by the linguistic community. Let me explain. Generally speaking, ideas are correct or adequate with regard to what they are supposed to represent. Now, what are ideas supposed to represent? On the one hand, simple ideas are adequate with regard to the properties that cause or can cause them. Pointing out the mere causal covariance however does not amount to a normative constraint. The normative dimension lies in the fact that the simple ideas are designed to represent certain things in a way that contributes to our biological and practical needs. Of course, in Locke this design is not to be taken in the contemporary evolutionary sense of an adaptive process but with regard to the assumption that God has established the relation between properties and ideas: «… simple Ideas … represent to us Things under those appearances which they are fitted to produce in us: whereby we are enabled to distinguish the sorts of particular Substances, to discern the states they are in, and so to take them for our Necessities, and apply them to our Uses.»35 Thus, the correctness or adequacy of conceptual content or general ideas is something that is governed by the epistemic agent’s cognitive make-up within the given environment. But this is not the whole story. On the other hand, there are the complex ideas made from these simple ideas. What are the normative constraints that govern them? With regard to substances, it might seem that they can or cannot conform to them. But since their real essences are completely out of our cognitive reach, there seem to be no standards of normativity available, except for the subjective ways our minds might abstract and apply them. Most commentators thus fall back on chiding Locke’s failed account of abstract ideas. In fact, however, Locke clearly points to another available standard, namely language. To cut another long story short, the normative dimension of ideational content crucially depends on the common use of language, since complex ideas are stabilised through their linguistic consolidation in the first place. Certain conceptual discriminations such as the distinction between water and ice or between gold and fools’ gold, for instance, may play a vital role in some cultures, while they might never take hold in other speech communities. Thus, the content of ideas tied to linguistic expressions is socially restricted: 34 See also M. Lenz, Lockes Sprachkonzeption cit., pp. 403-436. 18 «’Tis not enough that Men have Ideas, … for which they make these signs stand; but they must also take care to apply their Words, as near as may be, to such Ideas as common use has annexed them to. For words, especially of Languages already framed, being no Man’s private possession, but the common measure of Commerce and Communication, ’tis not for any one, at pleasure, to change the Stamp they are current in; nor alter the Ideas they are affixed to …»36 Differences drawn between objects or classes of objects, then, do not only depend on the causal impact of the world on our cognitive faculties. Accordingly, the normative question whether a certain ascription or differentiation is adequate or correct is not ultimately determined by what is given in nature, but depends on whether these ascriptions or differentiations do or will play a role in a certain linguistic community. While rooted in the pragmatic constraints of our make-up as designed by God, it ultimately is our cognitive development within a given linguistic community that governs the normativity of ideas. Again, Locke takes all of this to be grounded in processes that are established and habitualized without our taking notice of them. He illustrates this point with a simple example37. If you read a text, what you in fact see are the mere letters or traces of ink. However, what you actually grasp are not letters or traces of ink. Neither do you begin to painstakingly connect the single letters. Rather, you instantaneously grasp meaningful units of language. Yet, the fact that you manage to do so is not owing to some sort of givenness; it is 35 36 37 See J. Locke, Essay cit., IV, iv, 4, p. 564. J. Locke, Essay cit., III, xi, 11, p. 514. See also his famous ice-water example in Essay III, vi, 13, p. 447 sq.: «But to return to the Species of corporeal Substances. If I should ask any one, whether Ice and Water were two distinct Species of Things, I doubt not but I should be answered in the affirmative: And it cannot be denied, but he that says they are two distinct Species, is in the right. But if an English-man, bred in Jamaica, who, perhaps, had never seen nor heard of Ice, coming into England in the Winter, find, the Water he put in his Bason at night, in a great part frozen in the morning; and not knowing any peculiar name it had, should call it harden’d Water; I ask, Whether this would be a new Species to him, different from Water? And, I think, it would be answered here, It would not to him be a new Species, no more than congealed Gelly, when it is cold, is a distinct Species, from the same Gelly fluid and warm; or than liquid Gold, in the Fornace, is a distinct Species from hard Gold in the Hands of a Workman. And if this be so, ’tis plain, that our distinct Species, are nothing but distinct complex Ideas, with distinct Names annexed to them.» Cf. M. Lenz, Locke as a Social Externalist, in J. Marenbon (ed.), Continuity and Innovation in Medieval and Modern Philosophy, Oxford University Press, Oxford 2013, pp. 53-67. See Locke, Essay cit., II, ix, 9-10, p. 146 sq.: «… a Man who reads or hears with attention and understanding, takes little notice of the Characters, or Sounds, but of the Ideas, that are excited in him by them. Nor need we wonder, that this is done with so little notice, if we consider, how very quick the actions of the Mind are performed: For as it self is thought to take up no space, to have no extension; so its actions require no time, but many of them seem to be crouded into an Instant. … Secondly, we shall not be so much surprized, that this is done with so little notice, if we consider, how the facility of doing things, by a custom of doing, makes them often pass in us without our notice. Habits, especially such as begun very early, come, at last, to produce actions in us, which often escape our observation. How frequently do we, in a day, cover our Eyes with our Eye-lids, without perceiving that we are at all in the dark? … And therefore ’tis not so 19 not the case that these meaningful units are just given. Your ability is rather due to the fact that you have lived through a cognitive development in the course of which countless processes have been habitualized and socially consolidated. So if you perceive a green apple, this perception could not be sufficiently explained by appealing to the presence of a green apple. For this kind of appeal would simply amount to a causal explanation. According to Locke, the epistemic achievement of grasping an apple is never the result of cognitive process that could be rendered in merely causal terms; rather, it is the result of an ascription, i.e. of an implicit judgment that already involves normativity and needs to be explained with recourse to the cognitive development within a certain linguistic community. Conclusion: Nature versus Normativity? If one wants to draw a very brief conclusion, one might say that the claim that Locke fell prey to the Myth of the Given is itself a myth. On the one hand, it is a historiographical myth in that it takes for granted a fairly superficial reading of Locke’s theory of ideas. Once we take into account the dual perspective of explanation, it becomes clear that Locke can accommodate the distinction between the natural and the normative without giving up the foundationalist intuitions that drive his account. On the other hand, it shows that Locke’s theory might have resources to provide a systematic alternative to the contemporary ways of avoiding the myth. In following a dual perspective on ideas or cognitive vehicles it points to a way of combining, rather than separating, naturalistic and normative aspects. At the same time, Locke’s approach, however sketchy it is presented here, also shows that his foundationalism has to accommodate serious constraints. The reflections on normativity suggest that Locke’s way of getting around the Myth of the Given commits him to a conceptual relativism according to which the adequacy of our thoughts mainly depends on socially accepted determinants and thus on culturally variable factors. Whatever there might be in nature, a true apple is what we call an apple. However, this does not pose a destructive threat. Although Locke grants the possibility that certain classifications display local differences according to culture, his position ultimately precludes any kind of global relativism. Why is that? Well, our classifications and ascriptions are indeed often due to pragmatic needs and conventions, but they are still constrained by the assumption that we are beings with a certain biological make-up and that we live in one shared world. To put this point more specifically, we might want to remember the analogy of the vending machine. Ultimately, the strange, that our Mind should often change the Idea of its Sensation, into that of its Judgment, and make 20 insertion of coins and their suitability for the machine is not determined by us but a matter of how the machine is constructed. To be sure, one might want to question Locke’s divine teleology behind this idea. The general approach, however, might be just as easily combined with a different theoretical backup, such as a biological theory for instance. Whatever foundation we chose, Locke’s approach can still cohere with realist intuitions and remain committed to the idea that reality offers resistance and friction. For the appeal to the epistemic perspective does not rule out the impact of the natural historical perspective. In Locke, natural causal and epistemological explanations do not form an opposition but a complementary gradation. To put the point in more Ciceronian or indeed McDowellian terms, our epistemological and semantic norms are not opposed to nature but form a kind of second nature that is grounded in and constrained by the first nature. Summary In the wake of Wlifrid Sellars’ philosophy, John Locke’s theory of ideas is often taken to fall prey to the so-called Myth of the Given. The main charge is that Locke appeals to passively received sense impressions to justify knowledge claims and ultimately confuses natural processes with normative conceptual ctivity. In this paper, I will argue that the accusations are founded on a faulty reading and that Locke’s account does indeed circumvent Givenism without having to abandon the foundationalist ambitions that drive his theory of ideas. I will begin by exploring the attractions and pitfalls of the Myth. Secondly, I will show how the Sellarsian objections can be launched against Locke’s theory of ideas. Thirdly, I will present my interpretation of Locke’s take on ideas and show how they fare in relation to the crucial features involved in the discussion of the Myth. By way of conclusion, I will discuss whether Locke’s way of avoiding the Myth limits his foundationalist approach. Key words Ideas, Myth of the Given, Locke, Sellars one serve only to excite the other, without our taking notice of it.» 21