Transcendental idealism
A few weeks ago I attended the Transcendental Idealism workshop in London, which was organised by the Transcendental Philosophy and Naturalism project. The workshop was truly excellent. The papers were great, the discussions fruitful and the workshop atmosphere was very congenial. There were five talks in total; one on Friday afternoon and four on Saturday. Some of the papers and abstracts are available online on the TPN webpage and there are summaries of the papers at the TPN blog.
An interesting question that came up in some of the discussions concerned the nature of transcendental idealism. While many accepted that the distinction between noumena and phenomena was a metaphysical distinction, there was some dispute as to whether or not transcendental idealism should be understood as a metaphysical or an epistemological question. McDaniel claimed that transcendental idealism is the thesis that we are ignorant of the mind-independent properties of objects. That is, he takes it to be an epistemological thesis that appeals to a metaphysical distinction about properties.
I tend to think that transcendental idealism is a metaphysical view that amounts to claiming that there is a distinction between noumena and phenomena. That is, it is a metaphysical theory that distinguishes between things that are transcendentally real and things that though empirically real are transcendentally ideal. This is particularly clear when Kant claims that it was unfortunate that he chose the term 'transcendental idealism' since it is liable to be misunderstood, and that it would have been better to label his system 'critical idealism' or 'formal idealism'. In a letter to J. S. Beck he claims that transcendental idealism amounts to the claim that the forms of intuition are ideal. This ideality of space and time seems for Kant to be definitive of transcendental idealism.
That transcendental idealism is a metaphysical view comes out particularly clearly when considering the relation between the Inaugural Dissertation and the first Critique. The metaphysical system outlined in the Inaugural Dissertation already endorses the ideality of space and time and it distinguishes between noumena and phenomena. At that time, however, Kant did not make any claims about our ignorance of noumena, but rather thought that knowledge of noumena was possible by means of the understanding.
The epistemological limitations were only introduced as a result of the discursivity thesis. Ignorance of noumena followed because Kant claimed that knowledge requires both intuitions and concepts. By means of the discursivity thesis, knowledge of noumena via the understanding was no longer possible. Sensibility and understanding must work together for knowledge, and since sensibility is limited to that which is in space and time and therewith transcendentally ideal, it follows that we can only know phenomena. That is, ignorance of noumena follows from the conjunction of transcendental idealism and the discursivity thesis.
An interesting question that came up in some of the discussions concerned the nature of transcendental idealism. While many accepted that the distinction between noumena and phenomena was a metaphysical distinction, there was some dispute as to whether or not transcendental idealism should be understood as a metaphysical or an epistemological question. McDaniel claimed that transcendental idealism is the thesis that we are ignorant of the mind-independent properties of objects. That is, he takes it to be an epistemological thesis that appeals to a metaphysical distinction about properties.
I tend to think that transcendental idealism is a metaphysical view that amounts to claiming that there is a distinction between noumena and phenomena. That is, it is a metaphysical theory that distinguishes between things that are transcendentally real and things that though empirically real are transcendentally ideal. This is particularly clear when Kant claims that it was unfortunate that he chose the term 'transcendental idealism' since it is liable to be misunderstood, and that it would have been better to label his system 'critical idealism' or 'formal idealism'. In a letter to J. S. Beck he claims that transcendental idealism amounts to the claim that the forms of intuition are ideal. This ideality of space and time seems for Kant to be definitive of transcendental idealism.
That transcendental idealism is a metaphysical view comes out particularly clearly when considering the relation between the Inaugural Dissertation and the first Critique. The metaphysical system outlined in the Inaugural Dissertation already endorses the ideality of space and time and it distinguishes between noumena and phenomena. At that time, however, Kant did not make any claims about our ignorance of noumena, but rather thought that knowledge of noumena was possible by means of the understanding.
The epistemological limitations were only introduced as a result of the discursivity thesis. Ignorance of noumena followed because Kant claimed that knowledge requires both intuitions and concepts. By means of the discursivity thesis, knowledge of noumena via the understanding was no longer possible. Sensibility and understanding must work together for knowledge, and since sensibility is limited to that which is in space and time and therewith transcendentally ideal, it follows that we can only know phenomena. That is, ignorance of noumena follows from the conjunction of transcendental idealism and the discursivity thesis.
posted by Bader at 7:06 pm
8 Comments:
Hi Ralf,
Jeremy Dickinson here. In the event you do not remember me, I am a grad student at Syracuse University. We only spoke briefly about transcendental idealism, but I wanted to talk more. I found out through Ernesto Garcia that you had TI blog.
I wanted to comment on this post. I agree with you that transcendental idealism is a metaphysical view and that it is, more specifically, a metaphysical view concerning the status of space and time, which from the transcendental vantage point (and dare I say, world) are ideal.
Now, I have only read this post, so I do not know more about your view. Is your view that Kant is a two-worlder? Or is your view (as Dave Lu has maintained) that there is the real or noumenal world of which we are ignorant and the world we represesent to ourselves via logical constructions?
So I guess I have two questions: (1) What is your view on TI (be as specific or as general as you like here)?
(2) How does your view capture the sense in which Kant is an empirical realist whilst being a transcendental idealism?
By the way, I have ordered the Van Cleve at your suggestion. A good friend of mine studies at USC with him.
Thanks for your time, and for this blog :).
Best,
Jeremy
Hi Jeremy,
it was very nice meeting you at Syracuse.
Here are a few short answers to your questions:
1. Regarding one-world v. two-world. I follow van Cleve in this respect which means that, broadly speaking, I defend a two-world view though it obviously depends on what is meant by a 'world'. In particular if an ontologically substantial reading of world is used, then I believe that there is only one world. Otherwise, two worlds. At least there are two domains of discourse, rather than two ways of discoursing about the same domain. Phenomena are intentional objects and logical constructs out of intentionalia which means that they have a different and lesser ontological status than noumena. What precisely this ontological status is and whether they amount to a 'world' is a tricky question.
Van Cleve notes that the virtual-object theory "does not give us a dualism of two sorts of existent. ... My interpretation is nonetheless dualistic in the following sense: the distinction between appearances and things in themselves is a distinction between two separate universes of discourse - not between two ways of discoursing about the same class of objects." (Problems from Kant, p. 150) I generally agree with this, though I believe that we might be able to get 'a dualism of two sorts of existent' if we are willing to countenance various forms of existence, in particular some kind of intentional (in)existence. This would imply that there are different ways for things to exist and that phenomena exist differently than noumena, rather than the former having no existence at all.
2. Regarding empirical realism + transcendental idealism. Empirical reality is intersubjectively real, i.e. it is noumenally grounded but subjected to subjective forms of intuition which are shared by all human subjects. This is combined with transcendental ideality precisely because the forms of intuition are subjective and ideal. The transcendental ideality that pertains to phenomena contrasts with the absolute reality possessed by noumenal objects which are in no similar way dependent upon subjective forms of intuition. (In my PhD thesis I use various supervenience relations to model the relevant grounding relations and to capture the differing ontological statuses of phenomena and noumena.)
I hope this makes things a bit clearer. Let me know whether you think this makes any sense.
Enjoy the van Cleve book!
Best wishes,
Ralf
PS. Please give my regards to Ernesto and Dave.
Hello again Ralf,
Thanks for the reply.
About your response in 1, the question I have is whether you take your two-world view, which you qualify to be about two spheres (worlds) of discourse, to be Kant's view? Or is this what Kant should have said, though perhaps he lacked the conceptual resources or clarity to say so?
The reason I ask is (and this is related to your response in 2) that Kant seems to want to say that the empirically real world, if you will, is not merely real in virtue of being intersubjectively shared, but because there is an objective correlate to our intuitions, which are *given* in experience, but we can only know this object as it is ordered by our human modes of perceiving and conceptualizing. This object is the transcendental object=x, as you surely know.
Of course this is not to make a definitive claim on this matter, but I tend to defend a view (following my teachers in this subject; one of whom is Ernesto) that has it that the empirical object is not an intentional object precisely because the content of an empirical object is given to us in sensation, which given our peculiar forms of perceiving seems to require objects that are amenable to such forms. Otherwise, how would noumena affect us directly such that it provives the matter for empirical objects?
Maybe some of this (at least prima facie) disagreement rests on the concept of an intentional object. If I have this concept right, then what you mean is that the empirical object (as a logical construct) is the representation of an object that does not have mind-independent reality; it is merely a thing in thought, or something like this. But if this is right, I fail to see how this can be empirical realism even if communal intersubjectivity supplements the intention inexistence of the object.
Ok, I should stop here so you can reply (if you like). I will send my regards to Ernesto and Dave and look forward to your reply.
Best,
Jeremy
Hello!
I have had into working condition a material system of/for noumenal knowledge. Qualia included.
www.scatteredpiecesof1dying.blogspot.com
Write me!
Thanks a lot for your questions and comments.
About your response in 1, the question I have is whether you take your two-world view, which you qualify to be about two spheres (worlds) of discourse, to be Kant's view? Or is this what Kant should have said, though perhaps he lacked the conceptual resources or clarity to say so?
I take it to be the most defensible version of transcendental idealism. I think that there is some textual support for attributing this view to Kant, but am fully aware that it clashes with much that he has to say.
The reason I ask is (and this is related to your response in 2) that Kant seems to want to say that the empirically real world, if you will, is not merely real in virtue of being intersubjectively shared, but because there is an objective correlate to our intuitions, which are *given* in experience, but we can only know this object as it is ordered by our human modes of perceiving and conceptualizing. This object is the transcendental object=x, as you surely know.
I do not want to equate empirical reality with intersubjectivity - more is required than just that. What matters is the ground of the intersubjectivity, namely that it stems from noumenal affection combined with shared subjective forms of intuition.
If you mean by there being an objective correlate to our intuitions that is given in experience, that the matter of sensation derives from noumenal affection, then we are in agreement. However, it seems to me (but please do correct me if I am wrong) that you want to say more than this, in particular that you take this correlation relation to be more significant and to underwrite certain epistemological claims. If you were to mean this, then I would disagree. I know the phenomenal object and I know that there is something atemporal and aspatial that grounds it. But for me this does not amount to knowing the atemporal and aspatial object as it is ordered by our forms of experience. What I know is a different object, namely the phenomenal object. Simply because there is some metaphysical connection between the two objects does not imply that I know one by means of the other though only in that way in which the latter is known to me.
Of course this is not to make a definitive claim on this matter, but I tend to defend a view (following my teachers in this subject; one of whom is Ernesto) that has it that the empirical object is not an intentional object precisely because the content of an empirical object is given to us in sensation, which given our peculiar forms of perceiving seems to require objects that are amenable to such forms. Otherwise, how would noumena affect us directly such that it provives the matter for empirical objects?
Why does the source of the content determine whether or not it is an intentional object? For me, the source partly determines the ontological status of the object insofar as it determines whether it is merely subjective, or intersubjective or objective, but it does not determine whether it is an intentional object or a real existent.
I am not exactly sure what you mean by your question: "Otherwise, how would noumena affect us directly such that it provives the matter for empirical objects?" I think that noumena affect us directly insofar as noumenal objects affect noumenal selves. Do you believe that it is the empirical object that affects us? If so, how do you avoid problems of double-affection?
Maybe some of this (at least prima facie) disagreement rests on the concept of an intentional object. If I have this concept right, then what you mean is that the empirical object (as a logical construct) is the representation of an object that does not have mind-independent reality; it is merely a thing in thought, or something like this. But if this is right, I fail to see how this can be empirical realism even if communal intersubjectivity supplements the intention inexistence of the object.
The empirical object is not a representation - rather it is a logical construct. More precisely, it is the logical construct out of all possible intentionalia of the object. Intentional objects are the objects of our representations, they are that which our representations represent. I take it to be the case that actual intentionalia are mind-dependent insofar as they are immanent to the act of awareness. However, possible intentionalia out of which phenomenal objects are constructed are not in this way mind-dependent, even though they depend on the forms of intuition. It is precisely because possible intentionalia are grounded in noumena and because they depend on shared forms of intuition that they are empirically real, but transcendentally ideal. They are not empirically ideal since they do not depend for their (in)existence on acts of awareness and they are not transcendentally real since they do depend on the subjective forms of intuition.
Regards,
Ralf
Hello again Ralf,
You write:
"Why does the source of the content determine whether or not it is an intentional object? For me, the source partly determines the ontological status of the object insofar as it determines whether it is merely subjective, or intersubjective or objective, but it does not determine whether it is an intentional object or a real existent."
I guess I am confused. I took you to be claiming that empirical representations were intentionally inexistent objects, which I thought meant that these objects do not have an objective correlate, i.e., a mind-independent correlate. I will check into this. It is likely that I am just unclear on the ontological status of an intentional inexistent object. So I think you are right to say that the source of the content does not determine whether the object is intentional; however, I thought the source of an object at least partly determines whether the object is an *existent* or *inexistent* object.
You write:
"I think that noumena affect us directly insofar as noumenal objects affect noumenal selves. Do you believe that it is the empirical object that affects us? If so, how do you avoid problems of double-affection?"
I think I can agree with you here, but (currently) only in part. It may be that noumenal objects do affect our noumenal selves (though I want to be tenatative about the nature of this affection), but since my two-world view is (seems?) more substantive than your view I think there is an empirical world full of objects that are knowable insofar as they conform to our forms of perceiving and conceptualizing. These objects are mind-independent, but as such are only cognizable as the transcendental object=x. There is also a noumenal world that gives rise (in some unknowable way) to the empirical world, and as such is the ground of the empirical world, but this ground is inscrutable because to know it would be to know how God created the world. We can think about this world--indeed we must think about this world in order to arrive at the completeness in theoretical explanation that reason demands (cf. Antinomies); but then practical reason justifies the existence of noumenal objects (cf. the Canon).
So, I am not sure what the problem of double affection is supposed to be--I could be overlooking something here. I take it that noumenal objects affect us insofar as they affect our noumenal selves and phenomenal objects affect us insofar as they affect our phenomenal selves. From the transcendental vantage point--and world--space and time are ideal simply because the noumenal world lacks spatiotemporality.
Thank you for the responses.
Best,
JD
Hi Jeremy,
Thanks for your reply.
I took you to be claiming that empirical representations were intentionally inexistent objects, which I thought meant that these objects do not have an objective correlate, i.e., a mind-independent correlate.
It depends on what you mean by an empirical representation. The notion of a representation is ambiguous between the thing doing the representing and the thing that is represented. I take that which is represented, i.e. the object of a representation, to be an intentional object. That which is doing the representing is in my opinion a noumenal representation (though there can also be higher-order phenomenal representations that are themselves the intentional objects of noumenal representations).
It is also important to distinguish between appearances and phenomena. The former are the intentional objects of particular representations, while the latter are logical constructs that function as correlates of the former. I take appearances to be mind-dependent intentional objects that have phenomena as their correlates. Phenomena are logical constructs that are mind-independent.
I think there is an empirical world full of objects that are knowable insofar as they conform to our forms of perceiving and conceptualizing. These objects are mind-independent, but as such are only cognizable as the transcendental object=x.
What is your understanding of the transcendental object=x? Do you take it to be an object, in particular a noumenal object? Or do you take the transcendental object to characterise our understanding of objects when considered in abstraction from forms of experience? Is it the empirical objects or the noumenal objects that are only cognisable as transcendental objects, on your view?
There is also a noumenal world that gives rise (in some unknowable way) to the empirical world, and as such is the ground of the empirical world, but this ground is inscrutable because to know it would be to know how God created the world. We can think about this world--indeed we must think about this world in order to arrive at the completeness in theoretical explanation that reason demands (cf. Antinomies); but then practical reason justifies the existence of noumenal objects (cf. the Canon).
So, I am not sure what the problem of double affection is supposed to be--I could be overlooking something here. I take it that noumenal objects affect us insofar as they affect our noumenal selves and phenomenal objects affect us insofar as they affect our phenomenal selves.
The important question is what the relation between the noumenal and phenomenal realm is. Do you think that the noumenal world gives rise to the empirical world and that the latter than progresses on its own without input from the former? Or do you think that the empirical world is continually sustained by the noumenal world, that every empirical fact is directly grounded in the noumenal world rather than us having to trace down the chain of dependence within the phenomenal realm to the beginning of the empirical world which then would have to be explained in terms of noumena? If you accept the former view, then no problems of double affection seem to arise. However, on the latter reading, there are problems about overdetermination etc. If the noumenal object affects the noumenal self and gives rise to an intuition, then what role can the empirical object play? It seems that we either have (i) overdetermination, i.e. the noumenal object as well as the phenomenal object somehow produce the same result, or (ii) the noumena do all the work, or (iii) noumena and phenomena are largely disconnected. Moreover, there are questions about downward causation: does my phenomenal choice affect noumena?
Best wishes,
Ralf
PS. I will be away from St Andrews for the next few days, so my response will be delayed a bit.
Hi Ralf,
Sorry for the time lag in getting back to your last comments. End of semester stuff has kept me too busy.
You write:
"What is your understanding of the transcendental object=x? Do you take it to be an object, in particular a noumenal object? Or do you take the transcendental object to characterise our understanding of objects when considered in abstraction from forms of experience? Is it the empirical objects or the noumenal objects that are only cognisable as transcendental objects, on your view?"
Let's see, I do not think the transcendental object=x is an object considered (conceived of) as an abstraction from all our forms of thinking. Nor do I think it is a noumenal object. Rather, it is a placeholder for all objects of possible experience. What is meant here is that the transcendental object=x is not itself an object, because being a transcendental object it is merely the general form of an object. As a condition of the possibility of experience, Kant argues that there must be an object given in intuition. Since he is concerned in KrV with the transcendental conditions for the possibility of experience, he only argues for the transcendental object=x, which, again, is just the general form of an object of possible experience. It is a placeholder for that which is to be unified and "categorized" by the pure concepts.
Continuing, you write:
"The important question is what the relation between the noumenal and phenomenal realm is. Do you think that the noumenal world gives rise to the empirical world and that the latter than progresses on its own without input from the former? Or do you think that the empirical world is continually sustained by the noumenal world, that every empirical fact is directly grounded in the noumenal world rather than us having to trace down the chain of dependence within the phenomenal realm to the beginning of the empirical world which then would have to be explained in terms of noumena? If you accept the former view, then no problems of double affection seem to arise."
It seems all I (and Kant) is warranted in saying is that the noumenal is the ground of the phenomenal. I do not want to be committed to there being a 1:1 correlation between the noumenal and phenomenal realms. Nor do I want to accept that the noumenal merely gets the phenomenal going. I think Kant wants to avoid a stance on this issue also for the sake of intellectual humility. I understand the potential problems that can arise (double affection, overdetermination)because of your last comment, but I think Kant is silent on the matter of the exact relation between noumena and phenomena, though there *is* one to be sure.
Whether this means I should take such a stance on such matters is another story. Do I think, contra Kant, that I can pronounce precisely on the relation between phenomena and noumena? I do not think I can. Especially is this so since I think noumenon are monads (according to Kant and, if I understand monads, according to me). This means that noumena are God, the soul, and freedom. I might think we can say more (know more) than Kant thinks we can with respect to such entities (and yes, I think there exist such entities: I have a robust ontology). On the other hand, I think I agree with Kant that we can cognize freedom (as a causa noumenon). Perhaps we can know with more precision the nature of this particular noumenon, though I am unsure.
Ok, I need to run. I am going to be driving across the country, so I will likely be out of contact--unless I hit some hot spots on the way and can get the WiFi so that I can write.
I will be working on transcendental idealism (and other Kant topics all summer). I hope to keep this dialogue going, especially after I begin reading the Van Cleve. Thanks for your thoughts, and post more :).
Best,
Jeremy
Post a Comment
<< Home