The distinction between phenomena and noumena plays a significant role in Kant's thought but much of his readership up to the present has reversed the significance of these terms in crucial respects.
To many, phenomena concern only that which is in our head while noumena are outside of our head and are thereby the really-real which we seek to know. Because of this, there is a tendency to be disappointed when our knowledge is (apparently) restricted to the phenomenal. This is correct: we cannot know things as noumena (i.e., in themselves). However, we should shake the notion that it makes sense to see this as a new limit to our knowledge rather than a clarification of what knowledge is already. To illustrate this I will characterize knowledge in a crucial respect, then I will show how phenomena and noumena relate to knowledge to see which is more suited to be considered knowing.
Knowledge is objective and so something known by me can also be known to others - at least granting that they have the same access. For example, I can know the weight of an object by placing it on a scale and anyone else can do the same as long as I do not withhold the object from them. To call something known, then, we must assume all knowers have the same access to the object in principle, for otherwise, this would not be able to be considered objective.
Now, let us consider phenomena and noumena to see which may best fit the criteria given above.
Phenomena are thoughts of objects mediated by their appearance. These appearances are the product of our sensible faculty and even if we must see the object as mediated by our sensible faculty we still recognize phenomena as resulting from objects affecting us in some way. A mediated access is still some access.
Next, we take up noumena. Noumena are not appearances and do not relate at all to our sensible faculty. As their name suggests, noumena are intelligibilia and are in our representation only as thoughts. If noumena are taken to relate to objects standing outside of all representation they will only have this relation to the extent we arbitrarily think it. Adding thought upon thought does not constitute any sort of access to an object outside of that thought.
From the above, we may see clearly why it makes sense to characterize phenomena as knowable and why this doesn't seem coherent to say regarding noumena. For we can comprehend that equal access to objects as phenomena entails that the object can affect all knowers via their sensible faculty. On the other hand, equal access to noumena would require something like equal access to the contents of our thoughts, or perhaps a pre-established harmony of these thoughts.
With this matter in hand, it will be beneficial to consider what is going on when we suppose, uncritically, that there is something more objective or significant to knowing noumena (i.e., things in themselves).
We have already seen that knowledge requires equal access to the object and so phenomena are suited to be called objects of our knowledge because our sensibility is a common access point to the object. Now it must be emphasized that the phenomena is not merely the appearance of the object but consists also in our thinking of the object through the appearance. If we divide the phenomena into its sensible and conceptual elements and look to just those that are conceptual then we are left with merely the thought of an object in general, that is, with noumena. Here we see that we can characterize noumena as our thought of objects so far as they do not appear to us, and this is one way Kant does characterize them. We also see how easy it may be to carry sensibly given predicates over to our thinking of the object and lose track of how these predicates were only able to be given by the senses.
As thoughts, we find that sensible predicates must conform only to the form of thinking which ultimately consists of avoiding contradictions. However, for these sensible predicates to have been given about the object they first must have been sensed by us and thereby appeared to us. Anything that appears must do so under the form of appearance just as all thought must under the form of thinking. Now, when we deal with these sensible predicates as mere thoughts we may lose track of the sensible conditions for their appearing, but these conditions still ground the significance of these predicates to the object whether we have lost track of them or not. By losing track of these conditions we risk converting conditions of appearance into conditions of thinking. For example, we may think that it would be a contradiction to have an object with no dimensions, but no contradiction can emerge from the mere lack of a predicate. However, an appearance of an object must have a dimension and for it to lack this would not violate the form of thinking but the form of appearance.
Without our recognition of the above confusion of conditions for appearance as conditions for thinking, we would find it difficult to preserve the difference between knowing objects and thinking them without contradiction. Further, we risk losing track of their being both aesthetic and logical form or the difference between these two entirely. This is what Kant refers to as the confusion of appearances as things in themselves. Kant seeks to remove this confusion which should not suggest a new limitation on knowledge but only an attempt to keep knowledge within its proper limits.
