Transcendental idealism is occasionally identified with formalistic idealism on the basis of passages from Kant's Prolegomena to any Future Metaphysics, although recent research has tended to dispute this identification. Transcendental idealism was also adopted as a label by Fichte and Schelling and reclaimed in the 20th century in a different manner by Husserl.
Perhaps the best way to approach transcendental idealism is by looking at Kant's account of how we intuit (Ge: anschauen) objects, and that task demands looking at his accounts of space and of time. Before Kant, some thinkers, such as Leibniz, had decided that space and time were not things, but only the relations among things. Other thinkers, including Newton, maintained that space and time were real things or substances. Leibniz had arrived at a radically different understanding of the universe and the things found in it. According to his Monadology, all things that humans ordinarily understand as interactions between individuals and all things that humans ordinarily understand as relations among individuals (such as their relative positions in space and time) have their being in the mind of God but not in the Universe where we perceive them to be. In the view of realists, individual things interact by physical connection and the relations among things are mediated by physical processes that connect them to human brains and give humans a determinate chain of action to them and correct knowledge of them. Kant was aware of problems with both of these positions. He had been influenced by the physics of Newton and understood that there is a physical chain of interactions between things perceived and the one who perceives them. However, an important function of mind is to structure incoming data and to process it in ways that make it other than a simple mapping of outside data.
The salient element here is that space and time, rather than being real things-in-themselves or empirically mediated appearances (Ge: Erscheinungen), are the very forms of intuition (Ge: Anschauung) by which we must perceive objects. They are hence neither to be considered properties that we may attribute to objects in perceiving them, nor substantial entities of themselves. They are in that sense subjective, yet necessary, preconditions of any given object insofar as this object is an appearance and not a thing-in-itself. Humans necessarily perceive objects as located in space and in time. This condition of experience is part of what it means for a human to cognize an object, to perceive and understand it as something both spatial and temporal. Kant argues for these several claims in the section of the Critique of Pure Reason entitled the Transcendental Aesthetic. That section is devoted to the inquiry of the a priori conditions of human sensibility, i.e. the faculty by which humans apprehend objects. The following section, the Transcendental Logic concerns itself with the manner in which objects are dealt with in thought.
Kant's observations from a logical and philosophical point of view are supported in modern thought by some empirical findings that go beyond the science available to Kant in his time, are not based on what might be called a Kantian ideology, and yet support Kant's conclusions on the grounds of novel discoveries. Kant argues, essentially, that incoming data must be organized into a form that human minds can process. At the dawn of the computer age it was assumed that robots would soon be capable of taking over for humans in many tasks. Unexpectedly, it soon became apparent that pattern recognition was not an easy goal to attain. The human brain seems to be hard wired for pattern recognition. A very telling indication of organic structures for pattern recognition came to light when researchers discovered that the image of a moving object in crossing the retina is processed right in the retina and sends an almost instantaneous message: "Movement!" to the brain. So "movement" turns out to be an automatic processing of raw incoming data into a special signal having immense survival salience to the organism. Kant had to be satisfied with examining the functions of the mind and teasing out the functional dependencies without much if any help being derived from observable physical mechanisms in the brain. The mind imposes structures on incoming data. In the case of the rope perceived to be a snake, the initial structuring must be abandoned. The snake disappears from consciousness and is replaced by a rope. In various ways other philosophies have maintained this useful distinction between what humans conceive to be present and whatever may really be there. Important schools of modern philosophy of science, a field from which Kant drew much, speak in terms of "models" or "convenient fictions" rather than asserting actual knowledge of reality.
Certain interpretations of some of the medieval Buddhists of India, such as Dharmakirti, may reveal them to be transcendental idealists, since they seemed to hold the position of mereological nihilism but transcendental idealists who held that their minds were distinct from the atoms. Some Buddhists often attempt to maintain that the minds are equal to the atoms of mereological nihilist reality, but Buddhists seem to have no explanation of how this is the case, and much of the literature on the aforementioned Buddhists involves straightforward discussion of atoms and minds as if they are separate. This makes their position very similar to transcendental idealism, resembling Kant's philosophy where there are only things-in-themselves (which are very much like philosophical atoms), and phenomenal properties.
Some of Schopenhauer's comments on the definition of the word "transcendental" are as follows:
Schopenhauer contrasted Kant's transcendental critical philosophy with Leibniz's dogmatic philosophy.
In The Bounds of Sense, P. F. Strawson suggests a reading of Kant's first Critique that once accepted forces rejection of most of the original arguments, including transcendental idealism. Strawson contends that if Kant had followed out the implications of all that he said he would have seen that there were many self-contradictions implicit in the whole.
Strawson views the analytic argument of the transcendental deduction as the most valuable idea in the text, and regards transcendental idealism as an unavoidable error in Kant's greatly productive system. In Strawson's traditional reading (also favored in the work of Paul Guyer and Rae Langton), the Kantian term phenomena (literally something that can be seen from the Greek word phainomenon, "observable") refers to the world of appearances, or the world of "things" sensed. They are tagged as "phenomena" to remind the reader that humans confuse these derivative appearances with whatever may be the forever unavailable "things in themselves" behind our perceptions. The necessary preconditions of experience, the components that humans bring to their apprehending of the world, the forms of perception such as space and time, are what make a priori judgments possible, but all of this process of compehending what lies fundamental to human experience fails to bring anyone beyond the inherent limits of human sensibility. Kant's system requires the existence of noumena to prevent a rejection of external reality altogether, and it is this concept (senseless objects of which we can have no real understanding) to which Strawson objects in his book.
In Kant's Transcendental Idealism, Henry Allison proposes a reading that opposes Strawson's interpretation. Allison argues that Strawson and others misrepresent Kant by emphasising what has become known as the two-worlds reading (a view developed by Paul Guyer). This —- according to Allison false —- reading of Kant's phenomena/noumena distinction suggests that phenomena and noumena are ontologically distinct from each other. It concludes on that basis that we somehow fall short of knowing the noumena due to the nature of the very means by which we comprehend them. On such a reading, Kant would himself commit the very fallacies he attributes to the transcendental realists. On Allison's reading, Kant's view is better characterized as a two-aspect theory, where noumena and phenomena refer to complementary ways of considering an object. It is the dialectic character of knowing, rather than epistemological insufficiency, that Kant wanted most to assert.