Argument from illusion (November 2003):
Through sensory perception (sight, sound, etc) we come to know of the objects in the world around us. But are we immediately aware of these objects? Most answer this question in the negative. It doesn't seem like we are immediately aware of external objects as they are; there is some complicated process which occurs.
To support this claim some philosophers draw on an argument known as the argument from illusion. This argument goes roughly as follows: Consider an optical illusion, such as the appearance of a stick in water as being bent when the stick itself is really straight. What is being seen directly is bent -- not straight -- so it can't be the stick itself which is being seen directly.
Now here's where things get more complicated. The straight stick isn't being seen (directly), but some thing must be being seen. What is it? Something non-physical, a mental entity such as an image. Sometime earlier this century a number of philosophers (most notably Moore, Russell, and Ayer) subscribed to a theory of sense data (or sensa or sensibilia). When one looks at a circular coin and sees an elliptical shape or has a hallucination of a dagger which isn't really there, the object of immediate perception is not an elliptical coin or a non-existent dagger; it is sense data.
As far as I can tell, the arguments for sense data are pretty fuzzy, and sense data is no longer fashionable. I think it's clear that the argument from illusion goes wrong and that the fallacy can be seen by looking at the language. Even though we say "It is raining", we don't think that this means there is an "it" which is doing the raining. Similarly, there seems to be no reason to think that there is some "it which is being bent or elliptical" just because we say that what we see is bent or elliptical. Further, if what we are seeing is something mental, surely it is not bent or elliptical; it doesn't seem to make sense to speak of mental entities as being bent, only as representing some thing appearing bent.
Still, the argument from illusion and the idea of sense data are interesting because they touch upon other philosophical questions. Since we can have non-veridical perceptions (illusions, hallucinations, mirages, etc), can we trust our perceptions? If we believe in sense data, we can at least trust our mental experiences (even if they bear no strong relation to external reality). What are the inputs to perception, and how do we come to see objects? What does it mean to say that some thing appears to be other than it is?