Thomas Reid & Naive Realism

Mar 18, 2003 12:45

Well, I remember last night when going to sleep that this essay was due this morning. So I woke up nice and early and threw it together. Not too great, but whatever.



Naïve realism fails to acknowledge certain plausible questions concerning the legitimacy of our perceptions in regards to how they correspond to actuality. This is, I believe, its greatest failing as a philosophy of perception. It is often invoked on pragmatic grounds. In fact, the Scottish philosopher Thomas Reid had argued that the representational theory of perception was inferior to naïve realism simply on the basis that it does not entirely conform to one’s common sense notions of the world.
Reid went on to criticize David Hume, an advocate of representationalism, on the grounds that Hume, for all his skepticism, did not conduct himself as a skeptic in his personal affairs. It will hardly escape anyone’s notice that this is a bad argument. Firstly, Reid is arguing ad hominem, he fails to attack the doctrines of representationalism itself. Even if it were legitimate to pursue the point that Hume was inconsistent in his lifestyle, it does not at all follow that representationalism is false. Secondly, Reid’s premise of Hume living inconsistently with his professed skepticism is simply a false proposition. Hume was not inconsistent; he had very good reason to believe in the ordinary day-to-day perceptions he experienced. Hume had every good reason to, within his own epistemic framework, to believe that the next time he imbibes his favorite wine that it will indeed taste the same.

Reid also asserted, contrary to empiricist principles, that there is such knowledge as intuitive knowledge. It was from this intuitive faculty that one is supposed to be able to discern good from evil, beauty from ugliness, and so on and so forth. Hume, of course, would have, I believe rightly, nothing to do with such nonsense. The reason for this is as follows. Let us suppose, for instance, that Reid was sincere in his assertion that he has this alleged intuition and that it does indeed exist. Does this prove his case that he can indeed perceive and, therefore, know goodness? No. Any intuitive sensation is nothing more than a descriptive fact. To say that a specific event y caused the intuitive sensation f is purely descriptive and not at all normative. Hume successfully demonstrated that one cannot deduce judgments of value from judgments of fact. If every time Reid witnessed a robbery he also apprehended an unpleasant sensation this would add nothing to the case of whether or not robbery is immoral. The most he could say is that his moral sentiment found such actions are abhorrent. If Reid, illogically, persisted in maintaining that his intuitive faculty did present him with the idea of goodness then he were be stripping the term ‘good’ of its normative force and making “The intuitive sensation f is good” into an analytic statement of descriptive fact.

Where does this leave Reid with all his objections to representationalism? Certainly in not a very good position. Perhaps his last possible objection will be found in the allegation that the representational theory of perception cannot adequately account for our beliefs in physical objects, other minds, and other common objects of awareness. This is, as I have alluded to above, false, but it is at least worthwhile to investigate why it is false. Reid in his ad hominem against Hume charged him with living inconsistently with his espoused skepticism. Hume, according to Reid, had not right to believe that the next time he sat down in his study that the chair would not vanish or that the sun will rise tomorrow. What Reid failed to understand is that Hume simply applied a different set of criteria to justify belief; Reid, on the other hand, seemingly had no criteria beyond the fact that he took his percepts to be entirely legitimate without question. Hume asked the question and answered it while Reid refused to even acknowledge the question. There are inductive proofs and, to a stronger degree, inductive natural laws to which gave all the possible credibility needed in the empirical realm to such propositions as “The sun will rise tomorrow” and “Physical objects persist and do not dematerialize randomly”. Hume, of course, philosophically would rightly grant that there are not logically necessary propositions - a future instance could falsify them - but they are, in all respects, as close to certainty as is logically possible. Hence, there is no lack of accounting for our beliefs in the ordinary world under representationalism. There is merely a reworking of what counts as proof to these beliefs. Reid’s arguments against the representational theory of perception are, therefore, entirely fallacious.
Previous post Next post
Up