The Web of Belief – A Reading Note

I started reading Quine’s – The Web of Belief this weekend. In it there is one part which goes like this:

What are observations? Some philosophers have taken them to be sensory events: the occurrence of smells, feels, noises, color patches. This way lies frustration. What we ordinarily notice and testify to are rather the objects and events out in the world. It is to these that our very language is geared, because language is a social institution, learned from other people who share the scene to which the words refer. Observation sentences, like theoretical sentences, are for the most part sentences about external objects. This is why they can enter into logical relations with scientific theory, confirming or refuting it. (italics mine)

Quine, I think, rightly points that what we ordinarily notice are not colors patches, noises, etc; but objects and events… I think that is true, and more than that – those are the things we first notice, and only then we notice the colors, forms, and so on. It is true that the sense-data assumption first comes to mind when we learn the bio-physical facts. Namely, we learn how we have a “bag” of sensory receptors, and that each of these is separately affected by something in our surrounding. So, it is very normal to assume that the starting point of our consciousness of things is grounded in the consciousness of something that correlates with this bag of sensory events. So, the thinking goes, what we are conscious of in some moment is sense-data, and everything else is then synthesized on base of these sense-data.
But even this picture might make sense on first look,  lot of empirical research on conceptual development (for a nice sum up see Keil, 1989) shows a “holistic to analytic” shift. Those facts I think go strongly against the assumption that what the kids are initially aware is some sense-data, and that then the awareness of things as wholes which include more properties, and complex concepts is synthesized.
In previous posts, I said that this is because of the salience of things around us (by “things” I mean objects, properties, events, etc…). The whole objects tend to be more salient than specific properties (you will become aware of a rabbit, but not of its fur).

As I read that paragraph Quine isn’t thinking that we can become aware of wholes (in time or space) in the perception directly, without some sense-data to back it up. He seems instead to distinguish the awareness of the objects, which are supposed to be in publically accessible space, from colors, feels, etc… which are supposed to be something personal. Because of this Quine encounters a problem – how can we include persistent things in our observation sentences:

That there are enduring bodies at all, behind the passing show of sensory appearance, is a point of physical theory-a rudimentary point, but still something beyond the observable present occasion.

His explanation of how this is possible strikes me as unnecessary, given that other empirical researches show how children are aware of the objects as persistent well before learning language. This post has an example, but for a lot of examples which go beyond simple persistence, you might want to check Baillargeon, R.- “The Object Concept Revisited” (I read it in the Concepts – Core Readings book).

UPDATE:I added a post to give the general idea of the tests done by Baillargeon.

9 thoughts on “The Web of Belief – A Reading Note

  1. I don’t think that Quine agrees with what you are saying here, at least not if what you really mean is that infants become aware of events and objects. For Quine the infant first starts out with purely mass concepts and then has to learn how to talk about individual objects much later.

    Also, I don’t think you get the problem for Quine right either. The problem is not that he distinguishes between the objects and sense-data. His point is that the existence of objects is a theoretical claim not something that we observe. What we call an object is really a collection of ‘time slices’ and it is language that collects these things inot groups that we call objects. Furthermore there is in principle. according to him, no ‘right’ way to collect these time slices (this is his radical translation/inscrutibility of reference/ontological relativism)…

  2. Thanks for the comment, and for helping me in understanding those things

    1.Yes, I’m saying that we become aware of objects and events first, and yes, I don’t think that Quine would agree on that because. as you say, he sees existence of objects as a theoretical claim (as he says in that second part I quoted). Anyway, i didn’t want to say that my view is same as Quine’s – probably I wasn’t clear enough in my writing.

    2.Seems that I misunderstood Quine there. I thought that he is saying that objects are theoretical not just because of the time slices (though he explicitly mentions them as a problem at the place I quoted), but also because the data from the senses is not “integrated” in any way. So, I guess I was reading something which wasn’t there :).
    Connected to this, you say that for him infant first start with purely mass concepts. Can you give an example of what are those concepts?

  3. Richard,

    Further explanation about 2…
    What I thought was… if we (according to Quine) aren’t aware of things and events, what is that of which we can be aware of in one given instant? So, I assumed that he is thinking of the data that comes to us in that given moment. If it is not things, and it is not sense-data in one instant… what is that Quine says we are aware of in specific instant? Are those the “mass concepts” you mentioned (I guess those are not concepts like water, milk, etc..?)

  4. A mass noun is one like ‘milk’, ‘water’ and etc. A count noun is one like ‘cow’ ‘dog’ ‘event’ ‘object’ etc. According to Quine we don’t get those until later when we learn how individuate objects.

    What are we aware of for Quine? Y9ou neglect a third option; not the object (that is a collection of time slices) not sense data, but why not the time slice itself? What we don’t get is the object through time, but object at a time might be OK (as a loose way of talking, in Chisholm’s sense)

  5. Ah, OK, I misunderstood you about what you were saying about mass concepts, I thought you are saying that for Quine we get those before we learn how to individuate objects.
    I guess I didn’t see the third option because I don’t buy it. Thanks for pointing to it!

  6. Hi Clark,

    Here how Quine tries to explain, as he says, the development of theory that there are enduring bodies…

    The special virtue of observation sentences is that we can in principle learn them by ostension as wholes, keyed as wholes to the appropriate observable occasions, before ever learning to link the component words to enduring bodies. “The cat is on the mat” can be learned ostensively as a unitary string of syllables in association with a certain range of possible scenes. All of us necessarily learned some observation sentences thus. Then, as we gradually caught on to the theory of enduring bodies, we came to treat some of the component words as referring to bodies. Learning by ostension, as a trained animal might, to associate whole observation sentences with appropriate patterns of stimulation, is a first indispensable step toward learning physical theory. We get on into the theory afterward, bit by bit, as we learn to dismember the observation sentences and make further use of their component words. It is to this primary, ostensive learning of observation sentences as wholes that physical theory itself owes its vital continuing connection with sensory evidence.

    If the conclusions of the cognitive researchers are right, I think that there is something wrong about what Quine says. He is seeing the ability to be aware of objects as dependent on learning of language (“Learning by ostension, as a trained animal might, to associate whole observation sentences with appropriate patterns of stimulation, is a first indispensable step toward learning physical theory.”), but seems to me that this doesn’t align well with the cog.research results.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s