In tracing the journey from the sense experience humans share at its most basic level with animals to the formulation of scientific theories that endeavour to represent and explain the world imagined or at least conceived as a whole, Quine identifies several early stages: the global stimulus, observation sentences, and reification. On close inspection, none of these stages looks like manifestations of an organism whose capacity to be aware of the world is identical with brain events or irritation of nerve endings – and this remains true even if we overlook the issue of the origin of intelligibility and its place in the natural world. His account of the cognitive ascent of man is, however, worth examining if only for its boldness and (for the most part) clarity.
For Quine, as for John Locke, nihil in mente quod non prius in sensu – nothing in the mind that is not first in the senses. Hence the platform from which the journey to science begins is “the global stimulus”, the subject’s total sensory experience at a moment. Its “physical analogue” is: “the class of all sensory receptors that were triggered at that moment; or, better, the temporally ordered class of receptors triggered during that specious present”.27 This notion sounds like an unexceptionable start, with the surroundings of the organism being transformed into the proto-world of a subject. Unfortunately, this is not so. We are offered nothing to account for that in virtue of which the distinct and multiple deliverances to, and of, the various senses are totalled to the contents of a globe. As Kant was aware, something must secure the uniting of certain visual experiences with other visual experiences (to create a visual field) and visual experiences with experiences from the other senses – distance senses such as hearing, proximate senses such as smell and taste, and the experience of the body itself. Something in addition has to integrate this registration of physical impingements on the body with memories, thoughts, and other elements of consciousness that make sense of what is present. As if that is not difficult enough, something has also to prevent this union of diverse impingements becoming mere “merging”: the distinct elements have to converge while at the same time retaining their separation in space and time and their distinct qualitative identity. The rain of impingements has somehow to be united in the conscious moment of a subject while at the same time retaining its multiplicity.
Whatever one might think of Kant’s response to this problem and his appeal to “the forms of sensible intuition” (see Chapter 3), he at least noticed it. Quine seems to have overlooked the fact that a proto-world could not be built up out of a kaleidoscope of tingles and flashes and bangs, without something to tether them to; at the very least an array of objects set out in space and time. What is more, if the sense experiences are to mark the beginning of sense-making, they must be recognisable in the light of previous experience. This implies a temporal depth which sensory experience at a moment – a physical instant – or even a specious present does not possess. This last point carries more weight when it is recalled what is central to Quine’s naturalism: namely, that sense experiences are identical with neural discharges (which he calls “neural intake”) or, more broadly, with a present passive physical state of the nervous system, such that each state of the mind corresponds to a state of the brain-in-the-body. Such physical events could not unite while retaining their identity nor have temporal depth.
Quine correctly highlights the early crucial role of joined attention in creating an intersubjective world, en route to objectivity. There is “a public harmony of private standards of perceptual similarity”28 – guaranteed to be the same in all parties by their being shaped by the shared forces of natural selection – so that the individual can externalize what is a private experience. This, however, takes us further into a mode of shared sense-making than is warranted. Quine claims that the human mode of sense-sharing – “observation sentences” – are analogous to apes’ cries and bird calls, being like them “in holophrastic association with ranges of neural intake”.29 Observation sentences are basic, in that their truth does not depend on another assertion. They are taken from direct observation, and can therefore be understood without further knowledge. Examples would be “The man is standing up”, “the sky is dark”. The dubiousness of the idea of such a sentence making a direct and discrete connection of the realm of nature with that of discourse is something we shall further discuss. For the present we note that Quine seems to retrofit the grammar of referential discourse to the grammarless reactions of beasts – reactions that do not have referents, or not of the kind or in the way that sentences do. An animal’s cry is not an assertion of a matter of fact. While Quine does not ascribe the subsequent developments – “observational predication” (“That lion is dangerous”) and “observational categoricals” (“When it snows, it’s cold”) – to non-human animals, his smooth passage from ape calls to human discourse, stitching together a shared world, is effected only by slipping past the fundamental difference between animal squawks and observation sentences. With this free gift, it can seem as if the next steps – predication and general statements – are as good as already taken.
Behind the specific flaws in Quine’s legerdemain is a broader failure to appreciate how human language belongs to a subject facing the world – much of which is either absent or, if present, is so in general form – not an organism neutrally wired into its material surrounds. What is more, sentences do not arise individually; nor are assertions stand-alone in this sense. They belong to a network, just as beliefs belong to a network – something that on other occasions he seemed very much aware of, when he shared Donald Davidson’s notion of the holism of belief.30 We shall return to both these points in Chapter 6.
Hereabouts we encounter a greater challenge to a naturalism that tries to incorporate sentences and the utterances that use them into the material world of causes, effects and stimuli. There is no adequate naturalistic account of the contents of propositions and even less of those elusive parts of nature or the human world that are scissored out to be the contents of propositions. Even if there were sufficient correspondence between the referents of a (spoken or written) sentence – namely, what propositions propose – and a piece or aspect of the material world that would make a sentence true, the latter would not be simply a self-defining slice or fragment of nature. If I say “It was chilly yesterday” it is easy to determine the conditions under which this is true but not, except via language, pick out that which makes it true – even less what makes it true as I say it today. Yesterday’s chilliness does not have sharp boundaries and, more to the point, even if it did, it would not be available to be picked out by non-linguistic means
We can see this profound disconnection between on the one hand propositional and on the other bump-into-able reality – the kind of items that animals, too, could take hold of – between the referents of sentences and that which is experienced through our senses, most clearly when we think of assertions that turn out to be true but are not (yet) true at the time that they are asserted. I am referring to so-called future contingents. The content of the proposition “There will be a sea battle tomorrow” is both more and less than any actual sea battle that occurs the day after the assertion. It is more, inasmuch as it postulates a general possibility which a whole class of sea battles could realize. It is less in that (the other aspect of its generality) it lacks all the countless specific features any actual sea battle must have – ranging from the outcome to the facial expression of Able Seaman Jones at 10:01am during the battle.31
There are, of course, obvious problems with attempting to “naturalize” statements about abstract objects, future- and past-orientated propositions, and general, conditional, and normative statements. In short, with pretty well everything we say, from the homely remark that it is chilly for the time of year to the claim that the universe began 13.8 billion years ago. We reach knowledge fundamentally different from sense experience – the supposed reference of Quine’s “observation sentences” – long before we reach high-level assertions. This is connected with something
essential to the most basic assertions, such as “The man is standing up”: the mode in which their contents are shared. While I may keep what I know to myself, knowledge is intrinsically public. It can be known by any number of people. It lacks the fundamental, intrinsic privacy of experience, including the experiences that form the basis of observation. The reason for this is important: it is connected with the difference between animal experience and human knowledge, a difference that Quine elides by talk of “observation sentences”. The experiences of animals may be complex but they are fundamentally different from that which is asserted in “observation sentences”.
The ability to share knowledge is reflected in the fact that it is expressed in type words; the inability to share an experience, by contrast, is reflected in the fact that it is composed of tokens rather than types. While I can share the fact that I have a pain, I cannot share the pain. When I say “I share your pain”, this is only metaphorically true. The pain in my back remains in my back, irrespective of my having communicated it to you by groans, grimaces, or utterances. To put this another way, the actual pain I feel at t1 is not available to be experienced by anyone else; the piece of knowledge corresponding to the fact that I feel a pain in my back does not “occur” at t1 or at any other time. All that occur at particular times are events such as grimaces, groans, and verbal complaints. It is this difference that lies at the heart of the possibility of weaving together a shared body of knowledge and forms the fabric of the community of minds that underpins and maintains the human world. To revert back to our earlier discussion, it forms the “semantics” in “teleosemantics”. While we may share a telos with animals (such as not dying, or eating enough, or finding a mate), we do not share the semantics.
Quine’s third step, reification, seems to be entering the scene rather late. He claims that at the stage of observation sentences (and even at the stage of “observation categoricals” that express general expectations of connections such as “When it snows, it is cold”), there is no reference to bodies or other objects. Observation categoricals simply assert concomitance or close succession of observed phenomena and reflect the (testable) expectation that arises out of this. Surely, one would think, they should come before, rather than after, observation sentences; objects must be presupposed in observation sentences – “The man is standing up” – and their like. Quine at least acknowledges how remote the idea of an object is from the experience of a stimulus and the inadequacy of a phenomenalist approach to building up objects out of “neural uptake”.32 He argues that objects are “posits” and the “leap to reification” is based on a linguistic shift:
We can almost get [reference to objects] with the observation categorical ‘Whenever there is a raven, there is a black raven’. But not quite. The crucial leap to reification of ravens can be achieved by changing ‘there’ to ‘it’: ‘Whenever there is a raven, it is a black raven’[…] ‘It’ posits common carriers of the two traits, ravenhood and black.33
The carriers are bodies – full-blown objects – and the pronominal construction achieves objective reference.
It seems highly unlikely that the transition from the experience of the clustering of phenomena to the sense of an object as the substrate underpinning those phenomena and accounting for their coherence and stability, could be captured by something as specific and far down the cognitive track as the creation of a pronominal hinge – it – to link the features of an object. The verbal cart is being placed before the thing-like horse. Equally topsy-turvy is this characterization of the role of reification: “We [see] reification at work forging the sameness of reference between clauses. That is what it is for. The function of identity is recurrence in discourse”.34 Things, surely, must be the other way round. Sameness of reference presupposes sameness of identity, rather than establishes it. A pebble was enduringly itself long before a speaker made repeated reference to it and put it in inverted commas, referring to it as “that pebble”.
Reification, is a serious business, not easily, even less reliably, built on the stimulation of nerve endings such as gives rise to sense impressions. At the heart of the notion of an object is of something that exists in itself and has properties distinct from, or at least not exhausted by, our experiences. It is an enduring entity that is available for any number of observers: it is a public body that is enduringly available to be experienced and re-experienced because it continues to exist even when I or indeed anyone else is not sensing it and persists over time.
Quine focuses on persistence over time and imagines the human race – represented by an emblematic cave man – discovering objects when he realizes that the bear he and his friend are chasing is the same bear as the one that killed their friend last year. This primal scene seems entirely fanciful. More damagingly, it overlooks the need for a prior establishment of, at the very least, the objecthood of the reifier and of his or her conspecifics for there to be a community in which (enduring) things become a shared or common currency of reference. A world picture that already includes bears, friends dead and alive, and “last year” surely presupposes that of enduring public, objects, existing independent of those who encounter them. My sense of there being a “last year” in which my friend was killed seems to require the prior existence of that which Quine supposes to be established by communication about it.
We shall return to objects in the next chapter. For the present we note they lie at the heart of the human intuition of the world as “out there” in the sense of being other than our selves; and this is a fundamental, global intuition that is inseparable from the full-blown “aboutness” or intentionality of our conscious experience. Objects are the cornerstone of our public world, and the godfather of all the different kinds of shared spaces that characterize that world. It seems difficult to imagine them being delivered by neural activity that is the mere effect of the impingement of energy on an organism. It seems not unreasonable, therefore, to conclude that Quine’s endeavour to collapse the gap between mind and nature creates: a) an insuperable barrier to explaining our everyday ontological commitments, from material bodies upwards; and b) his need to overlook intentionality even in cognitively advanced items such as “observation sentences”. Reification which “proved indispensable in connecting loose ends of raw experience to produce the beginnings of a structured system of the world”35 eludes Quine’s explanation. It is certainly indispensable but there is nothing in the rickety structures of his naturalized epistemology to begin to account for it.
Undaunted, Quine presses bravely onwards on the journey from tingles to theories and arrives at science which he characterizes as “a biological device for anticipating experience”.36 The most fundamental aspect of science is the scope and robustness of its general assertions, facilitating such anticipations, although that which is anticipated may be something as esoteric as the conditions under which we can technologically exploit certain mechanisms in natural processes or the outcome of an experiment. Quine’s jump from the global stimulus to observation sentences seems to smuggle in the capacity for generalization – predication and categorization – as a faculty not requiring explanation, rather as with Millikan’s “chains of reasoning” referred to earlier. Seen clearly, however, the human capacity for generalization is an extraordinary faculty that cannot be taken for granted and if understood correctly cannot be naturalized.
General assertions allocate objects and events to classes. True classification involves seeing an entity X as both its (token) self and as belonging to a type. This duality of cognition is linked with many other capacities or operations. They include:
a)
distinguishing two levels of identity of an entity – type or qualitative and individual or numerical identity – such that an object may be both an instance of a class and a particular occupant of a location in space and time;
b)
reclassifying an entity under more than one type – seeing that item over there as a chair, a piece of furniture; an obstruction; a
fire-hazard on account of its being inflammable; a treasured heirloom; and so on, without apparent end;
c)
separating the entity from its meaning or significance on a given occasion – as when we realize that something that is a nuisance is also a chair;
d)
on the basis of this, communicating intelligence about the entity.
If it is worth spelling out these rather obvious points, it is because they highlight the fundamental difference between human classification of the contents of the world and what is called “classification” as something carried out by animals. The capacity to classify items is ascribed to animals on the basis of the failure to discriminate between items that share key general characteristics. A male robin may respond with aggression to all objects that have a combination of a large red spot on a brown background – a stuffed red breast, for example – as if it were a rival male.37 This is called “stimulus generalization” and is seen as a primitive or even primordial form of classification. In fact, it is a mistake to see such failures of discrimination as evidence of the presence in the creature’s mind of a broad class of items called “aggressors”. Moreover, the classes identified in the case of so-called animal classification – which seem to aggregate items that elicit the same responses – are not distinct from the objects or events that fall under them. Humans separate the classified members from their class. It is this distance between individuals and the classes to which they are ascribed that permits reclassification, most evident in full-blown predication that explicitly predicates the class of the entity seen as a subject: “Socrates is a man”, “Socrates is a bore”, “Socrates is a Platonic myth”. Indeed, I can refer separately to classes as I do when I use common nouns such as “table”, “cloud”, “tree”.
Logos Page 14