We live and move in space, here and there and everywhere, and this space seems in many ways to be an integral whole. If one attempts to consider the question of what constitutes our first basic experience of space, however, it turns out that things are not as simple as they may seem.
The locus classicus for this point is Berkeley's New Theory of Vision, in which he argues that we do not, strictly and properly, see distance but learn to associate distances with certain sights, given other things like the muscle-pressure of the eyes or the degrees of confusion in how objects appear. We see instead extensions of color. Since we don't perceive distance by vision as such but only by visual experience, there must be some other source, and Berkeley locates this in the sense of touch. When I see something, these are so closely connected to certain tangible sensations that we get when moving around that the one suggests the other automatically. If we think of this as visual space, visual space is not itself spatial but a sort of ancillary apparatus for thinking about tangible space, the space we think of ourselves as moving around in.
If this is the case, though, vision does not provide the only ancillary apparatus for thinking about space; obviously hearing does as well. This is perhaps most dramatically captured by the Doppler effect of a screaming whistle flying by, but all moving objects convey this auditory space. Berkeley himself recognizes this (NTV 46-47), although he does not dwell on it at great length. We also hear things in front of us, behind us, to the side, far and near. Berkeley suggests that the connection between the ideas of sight and the ideas of touch is much more intimate than that between the ideas of hearing and the ideas of touch -- we are less likely to think of sounds as 'in' the things we touch than we are to think of colors as 'in' them, for instance -- so that there is a kind of hierarchy here, in which touch gives us distance, sight gives us a strongly connected and very rich apparatus of signs (so much so that Berkeley thinks one can consider it a language), and hearing a more loosely connected and less rich apparatus of signs, in which the signs signify the tangible.
(One might ask if smell and taste, for instance, have their own space in the same sense we care considering here. One might think there's a case for taste; one tastes strawberry in one's mouth. The spatiality is clearly not that of the taste itself but of the touch, e.g., the tangibility of our tongues and the roofs of our mouths; but if Berkeley is right, this is true of sight and hearing, as well. Hume gives an example of putting a fig at one end of a table and an olive at another; we classify, so to speak, the taste of the fig with the fig's end of the table rather than the olive's end. One could take this as an argument for taste as another ancillary apparatus, although more loosely connected than hearing, and smell as perhaps even more loosely connected.)
Berkeley takes touch to be the fundamental ground here in part because of his strong empiricism; ultimately a spatial relation like distance must reduce to some sensible idea of some specific character; sight and hearing, in the kinds of contexts in which we say we see or hear distance, give us ideas that seem much more obviously to be functioning as signs of other ideas than any of the ideas of touch. But if we relax the empiricist assumption, it becomes clear that touch ends up as a fundamental because there's nowhere else to go if you are strict empiricist. If you are not a strict empiricist, it becomes possible to take tangible space also to be an apparatus of signs. Then phenomenal space, space as it appears to us in our everyday lives, becomes an interrelation of three different systems of signs, three different spaces that are capable of functioning as signs of each other; and the whole appearance of things, tactile, visual, auditory, is itself a sign of something not sensed but understood (something like this is Lady Mary Shepherd's view), or perhaps something recognized more fundamentally in our integration of these different kinds of sensation in the first place (as in views that posit an internal sense, like a sensus communis, capable of recognizing space).
