More on the Riddle of Intentionality with the Help of Molnar

  According to George Molnar,

The fundamental feature of an intentional state or property is that it is directed to something beyond itself . . . All mental states and processes have an internal reference to an object. The identity of the intentional state is defined in terms of this intentional object. . . . Since intentionality constitutes the identity of mental phenomena, it follows that the nexus between the mental state or process in question and its intentional object is non-contingent. (Powers: A Study in Metaphysics, Oxford, 2003, p. 62, second and third emphases added.)

Molnar is right: the directedness beyond itself to an object is an internal feature of the intentional state. Consider an act (intentional state) of seeing a particular green paloverde tree. What makes the mental act a consciousness of that very object? Some will be tempted to say that the tree in reality, outside the mind, causes the mental state both to be directed and to be directed  to the very object  to which it is directed.  But then the object-directedness would not be an internal feature of the intentional state.  The curious thing about the nexus of intentionality is that mental acts are intrinsically directed to their objects.  They refer beyond themselves by their very nature. So it is not in virtue of an external relation to an external thing that a mental state is object-directed.  As I argued earlier, object-directedness is not to be confused with object-dependence. One should not allow the prevalence of various forms of externalism over the last 35 years or so to blind one to the predominance of internalism in intentionality theory from Brentano on.  (This is not to say that there are no object-directed states  whose identity does not require the existence of an external referent.)

If one were to suppose that the object-directedness of every act requires the existence of external things, then (i) there would no object-directedness in the case of acts directed to nonexistent objects such as the merely possible golden mountain and the impossible round square, and (ii) an intentional state would lose its intentionality should the external thing to which it is directed cease to exist.  In the case of (i), what either does not or cannot exist cannot do any causing, and in the case of (ii), what no longer exists cannot do any causing either.

Consider again my Washington Monument (WM) example. If, unbeknownst to me, it ceases to exist while I am merely thinking about it, but not sense-perceiving it either directly by ordinary vision or indirectly via television, the directedness (intentionality) of my thinking is in no way affected by the WM's ceasing to exist: my conscious state remains directed, and it remains directed to the very object to which it was directed, and indeed in exactly the same way, say, under the incomplete description 'monolithic marble obelisk.' But what object is that? Which object is the intentional object? Is it the transcendent WM itself? Or is it an immanent object? There is a puzzle here that cannot be solved  by stipulative definition of 'intentional object.' Two possibilities.

P1. One possibility is that the intentional object (IO) is the WM itself.  There is good phenomenological reason to maintain this. After all, when I think of the Washington Monument, my thinking is directed beyond itself to something other than itself: I am not thinking about some intermediary item or epistemic deputy or surrogate such as a sense datum, idea, image, way of being appeared to, representation, guise, noema, or whatnot.  My thinking goes straight to the transcendent thing itself; it does not stop short at some immanent item that plays a mediating role.  It seems we ought to say that the IO is the transcendent thing itself.  

If so, the WM is my act's IO both while the WM exists and after it ceases to exist.  Don't forget that it is a phenomenological datum that the IO remains self-same over the interval despite the fact that during that interval the WM ceases to exist. Now the WM is in no way immanent to consciousness; it is neither a real content thereof in Husserl's sense of  reeller Inhalt, nor is it immanent in the manner of an Husserlian noema.  No wholly determinate 550-foot-tall marble obelisk resides in my head or in my mind. It cannot be in or before my mind because my mind, and yours too, is finite: it cannot 'wrap itself around' the entirety of the massive monolith. Only a tiny fraction of the WM's parts, properties, and relations are before my mind when I think of it.  That would also be the case were I standing in front of the monument looking at it.

So on (P1), the WM is the IO of my act, and the WM, both before and after it ceases to exist, is one and the same transcendent  item.  After it ceases to exist, however, it is a nonexistent transcendent item without ceasing to be the IO of my act of thinking.  That is to say: my ongoing thinking  of the WM has available to it an IO over the entire interval, an IO that has and then loses the property of existence.  Note the difference between 'My thinking has no object' and 'My thinking has an object that lacks existence.'

(P1) thus lands us in the Meinongian predicament of having to affirm that some items are both transcendent of consciousness and thus in no way mind-dependent, and without existence. (I am assuming the untenability of any distinction between being and existence; hence there is no escape by this route.)  I will say that an item that has neither existence nor being of any sort is 'beingless.' It is a pure 'what,' a pure Sosein bereft of Sein. It is ausserseiend. 

I myself find the notion that some items are beingless unintelligible although I do understand how the notion is arrived at. Some will dismiss my finding of unintelligibility as a merely autobiographical remark, but by my lights it is more than that.  It just makes no sense to say that there are, in an ontically unloaded or non-committal sense of 'there are,' definite items actually possessing properties and thus numerically different from one another that are both transcendent of consciousness and jenseits von Sein und Nichtsein, "beyond being and nonbeing." 

Therefore, while there is good phenomenological reason to maintain that the intentional nexus puts us in touch with  the thing itself and thus that the intentional object of an act is the thing itself, this plausible view entangles us in seemingly  insuperable Meinongian difficulties.  My thinking of the WM does not become objectless  half-way through the interval. That is phenomenologically obvious. Therefore, if the WM  is the IO of my act, then the WM becomes a nonexistent object by the end of the interval.  As I noted earlier, Husserl in the 'Jupiter' passage in the Logical Investigations seems headed in a Meinongian direction.

We face a serious problem if Meinongianism is to be avoided. We want to say that in every case intentional states are directed to things themselves and not to immanent intermediaries.  We want to say that the IO is the real thing 'out there in the world.'  But the problem of nonexistence (not inexistence! pace so many historically ignorant analytic philosophers) throws a spanner in the works.  One could say, and it has been said, that when the IO exists, the act gets at it directly; when the IO doesn't exist, the act terminates at a representation in a mind.  This is an option that needs discussing in a separate post.  For now I am assuming that in every case, the IO is either a transcendent item or an immanent item. I have argued that on the first alternative the upshot is Meinongianism, an upshot that by my lights is unacceptable. 

P2. The other possibility (theoretical option) given the assumption just stated is that the IO of my ongoing act  of thinking of the WM during an interval in which it passes from existence to nonexistence is not a transcendent item, but an immanent item.  Two sub-possibilities (theoretical sub-options) suggest themselves.

P2a.  On the first sub-option, the IO is a representation R in the mind.  To say that the IO exists is to say that R represents something in the external world. To say that the IO does not exist is to say that R does not represent anything in the external world.  So when I am thinking about the WM, during the entire time I am thinking about it, what I have before my mind is a representation R_wm which at first represents something and then ceases to represent anything but without prejudice to its being one and the same representation during the entire interval. This suggestion accommodates the fact that, phenomenologically, nothing changes during the interval.  But it succumbs to other objections. Husserl fulminates against representationalism and its notion that consciousness is like a box with pictures in it of things outside the box.  See Husserl's Critique of the Image-Theory of Consciousness.If an intentional state is directed to what is beyond itself, as Molnar rightly states above, then it is not representations to which consciousness is directed, but the  things themselves.

P2b. On the second sub-option, the IO is an immanent item, but not  a representation. It is an ontological 'part' of the thing itself.  Suppose the tree I see is a synthetic unity of noemata.  The transcendence of the tree is constituted in the potential infinity of the series of noemata, but each noema is inseparable from a noesis. This leads to idealism which is arguably untenable. But I cannot say more about this now. 

The intentional nexus as non-contingent

Molnar tells us above that the link between act and object is non-contingent. The reason is that acts are individuated by their objects: every act has an object, and what makes an act the act it is is its object.  Since an act cannot be without an object, an object that makes it the very act it is, the nexus between act and object is non-contingent.

But if in every case an act cannot exist and be the very act it is without an object, then, if the external thing does not exist,  as in the case of the Roman god Jupiter, the object must  be a Meinongian nonexistent object.

The intentional object may or may not exist

"The intentional object can be existent or non-existent." (Molnar, 62) He infers from this that the intentional relation cannot be a genuine relation given that a genuine relation cannot obtain unless all its relata exist.

But we should note an ambiguity in Molnar's formulation. The formulation uses the modal word 'can.' But is the point non-modal or modal? Are we being told that some IOs exist and some do not? Or that every IO is such that, if existent, then possibly nonexistent, and if nonexistent, the possibly existent?   I should address this in a separate post.

We should also note the following. If the intentional nexus is not a relation (because some IOs exist and some do not), and the act-object nexus is non-contingent such that, necessarily, every act has an intentional object,  then in the cases where the IO does not exist, and Meinongianism is false, the IO must be an immanent object.  So at least some IOs are immanent objects given the internality and non-contingency criteria cited by Molnar.  But if some IOs are immanent, then the pressure is on to say that they all are, which leads us either to representationalism or to transcendental idealism, both of which are deeply problematic.

The indeterminacy of intentional objects

Finally, among the non-linguistic criteria of intentionality, Molnar mentions the fuzziness or indeterminacy of intentional objects (p. 62). It is clear that some intentional objects are, as Molnar says, "seriously indeterminate." Suppose that I am expecting a phone call soon.  To expect is to expect something. The object expected, the phone call,  is indeterminate with respect to the exact time of its arrival. It is indeterminate with respect to other properties as well. But is every intentional object indeterminate?  The WM exists, and whatever exists is wholly determinate.  But when I think of it or remember it or expect to see it or perceive it, what is before my mind is not the WM with all of its parts, properties, and relations. Given the finitude of our minds, it would be impossible to have the whole of it before my mind. The WM, precisely as presented, cannot be the WM itself.  The former is indeterminate in many but not all respects whereas this is not true of the latter.  What this suggests, given the internality and non-contingency criteria is that the intentional object is not the thing itself, but an immanent object.  

Aporetic conclusion

We want to say that in every case intentional states are directed to things themselves and not to immanent intermediaries.  It is a phenomenological feature of intentional states that they purport to reveal things that do not depend for their existence on consciousness.  My visual perception of  the tree in my backyard purports to make manifest a thing in nature that exists and has many of the properties it has whether or I or anyone ever perceives it.  That purport is built into the phenomenology of the situation. We therefore want to say that the IO is the real thing 'out there in the world.'  But then we bang up against the problem of intentional nonexistence.  

We seem to face a dilemma. Either the IO  is the thing itself or it is not. To hold to the identity of the IO and the thing itself, we must enter Meinong's jungle. We have to embrace the unintelligible notion that there are transcendent nonexistent items in those cases in which the IO does not exist. On the other hand, if we hold that the IO is an immanent item, then the problem of its relation to the thing itself arises. Is the IO a representation of the thing itself? Or is it an ontological part of the thing itself? Either way there is trouble.

 

Comments

[Dave Bagwill]

It is my intention (!) to understand this entire thread; but, while I have followed it - with difficulty - I don't understand it well enough to explain it to an 8-year-old. Which is the bar some people have established as a measure of whether one really understands a topic.
(A bit of tongue-in-cheek there.) I am enjoying tagging along.

[BV]

Dave,

I would be happy to try and answer specific questions. NOT: what are you saying? BUT: what do you mean by 'immanent'?

You will have gathered, I hope, that your intention to understand these posts is an instance of intentionality as philosophers use this term, but that intentionality covers mental states that in ordinary language we would not call intentions, e.g., my noticing that a cat has entered the room.

[Dave Bagwill]

Thanks Bill. I am keeping up; I read Pierre Jacob's entry at the SEP for a little background, since I was unaware of the arena that you and others were doing battle in.

[oz]

Actually, Molnar does give something like a definition.

All mental states and processes have an internal reference to an object. The identity of the intentional state is defined in terms of this intentional object.

However, he doesn’t define ‘internal reference’. So logicians are now asking what justifes the inference from (A) to (B) as follows:

(A) Jake is thinking of Zeus
(B) Jake’s mental state has an internal reference to an object, and the identity of this mental state is defined in terms of this intentional object.

Presumably the mental state in question is what is predicated by ‘thinking of Zeus’? OK. But how does that show that there is some object, an intentional object, to which there is internal reference? Where is the minor premiss that gets us from (A) to (B)?

[oz]

>I don't understand it well enough to explain it to an 8-year-old

Not difficult, if the 8 year old has the patience. Take him/her/it through the following moves.

(1) Suppose Jack is thinking of Santa Claus

(2) Then Jack is thinking of someone.

(3) So there is someone Jack is thinking of

(4) But there is no such person as Santa Claus

(5) So there isn't someone Jack is thinking of

It's important that they are 8 years or older, otherwise they will disagree with (4)!

[Dave Bagwill]

Thanks oz. Though, in reading that entry at SEP, even a smart 9 year old would have some trouble solving the following, which I think is still the problem? :
"Now, the full acceptance of Brentano’s first two theses raises a fundamental ontological question in philosophical logic. The question is: are there such intentional objects? Does due recognition of intentionality force us to postulate the ontological category of intentional objects? This question has given rise to a major division within analytic philosophy. The prevailing (or orthodox) response has been a resounding ‘No.’ But an important minority of philosophers, whom I shall call ‘the intentional-object’ theorists, have argued for a positive response to the question. Since intentional objects need not exist, according to intentional-object theorists, there are things that do not exist. According to their critics, there are no such things. "

[BV]

OZ,

In your response to Dave, you make the mistake of thinking that (3) follows from (2). It obviously doesn't given your existentially loaded use of 'some.' The move from (2) to (3) is not like the move from

2* Jack is pounding on someone

to

3* There is someone Jack is pounding on.

[BV]

As for your response to me, you mistakenly assume that (B) is being inferred from (A). Not so. (B) is merely a plausible analysis of (A). You have this crude idea that thinking about something is standing in an external relation to it. Which is to say that you do not understand the phenomenon of intentionality.

And that explains why, in your book, you mistakenly assimilated object-directedness to object-dependence.

[BV]

Dave,

I take it you don't understand what is being said in the SEP passage. Let me see if I can simplify it for you. I will try to avoid technical jargon such as 'act,' 'intentionality,' immanent,' 'noema,' etc.

Would you say that imagining a unicorn is the same as or different from imagining a leprechaun? They are different even though in both cases there is an imagining. Agree? Now what makes them different? There are has to be something that makes them different, for they plainly are phenomenologically different even though they are both instances of imagining. They don't differ as imaginings so some further factor must brought in to explain the difference. Two more obvious points. First, one cannot just imagine. Agree? Second, in my examples the imaginings are directed toward items that do not exist. Agree? By 'exist' I mean exist in reality outside the mind. And note that I am assuming that a concrete episode of imagining is a state of mind or a mental state. (That's not obvious, but it is very plausible.) I'll stop here and wait to see if you find all of the foregoing crystal clear before proceeding.

[Dave Bagwill]

Thanks Bill. I actually thought I DID understand that passage from SEP. Did it not lay out the fundamental problem that leads to different approaches discussed/argued in this thread, especially twixt you and London?
But I am enjoying your walk-through, please continue! I have crystal clarity with your explanation thus far.

[BV]

Dave, to continue:

There has to be a distinction between an imagining and something else. But what is that? One naturally says that to imagine is to imagine something. But here is where London balks or rather baulks. For he thinks that if Jake imagines something, then there exists something that Jake imagines. He thinks that imagining must be like eating. To eat is to eat something, but if one eats something, then there exists something that one eats. One cannot eat what does not exist. And so London thinks that if Jake's imagining is directed to an intentional object, that object must also exist -- which contradicts the fact that unicorns and leprechauns do not exist.

One way out is to say that there are transcendent items that do not exist (or have any mode of being) and that such intentional objects as unicorns and leprechauns are among them. More later if you like.

[Dave Bagwill]

Thanks again, Bill. I will now go back over the thread and see what I can see.
Whenever the word 'exists' enters into an argument, I've come to automatically anticipate the same sort of differences.

[OZ]

>In your response to Dave, you make the mistake of thinking that (3) follows from (2).

Of course it doesn't follow. It involves the 'intentionalist fallacy', namely of moving from intentional to non-intentional contexts. My point is that the ‘paradox’ simply dissolves once we understand the fallacy for what it is.

>As for your response to me, you mistakenly assume that (B) is being inferred from (A). Not so. (B) is merely a plausible analysis of (A).

Here are the two propositions again.

(A) Jake is thinking of Zeus
(B) Jake’s mental state has an internal reference to an object, and the identity of this mental state is defined in terms of this intentional object.

So you are not ‘inferring’ (B) from (A). But then, supposing (A) is true, what possible reason is there for supposing (B) is true? Over to you.

>You have this crude idea that thinking about something is standing in an external relation to it.”

Where did I say that thinking about something is standing in an external relation to it? It’s the exact opposite of what I hold.

> For he [Londoner] thinks that if Jake imagines something, then there exists something that Jake imagines. He thinks that imagining must be like eating.

Blimey (as Londoners say). That is the exact opposite of what I say. What I say, see my guest post below, is that ‘imagines’ is an intentional (or ‘non-existential’) verb, so we cannot infer from “S imagines something” to “there is something imagined by S”.

[OZ]

“Such mental states refer beyond themselves to objects that may or may not exist” (Vallicella, link).

Molnar: “The fundamental feature of an intentional state or property is that it is directed to something beyond itself”

Both these claims use the term 'beyond'. Bill, do you agree with these statements?

[David Brightly]

Morning Bill.

To my mind, this piece and others of yours on this topic amount to a successful reduction of the concept 'intentional object' to absurdity. The idea contains too much to be self-consistent. What do we throw away? The obvious answer to me is to retain 'intentional' and discard 'object'. We keep the directedness but drop the idea that it must terminate on a thing of some sort. Consider Max Black's iron spheres. The tuple [iron, sphere] has a direction in 'conceptual space'---the two properties iron and sphere are somewhat analogous to the components of a vector---but it's wholly contingent whether it 'points to', ie, describes none, one, or many actual objects. Likewise, if I know nothing of the WM but you tell me that there is a monolithic marble obelisk in the USA called the WM, then I can rehearse the thought that [monolithic, marble, obelisk, in USA] is instantiated in the WM, but I don't at all get the sense that this thought uniquely 'reaches' or 'goes straight to' the Washington Monument. For that something more must be added. The 'objectuality' must be put back in. I think this extra factor must be some degree of personal acquaintance with the WM, either directly through the senses, or failing that images consistent with the description, or augmented description that would enable sensory acquaintance in principle. The intentional arrow must be fleshed out with real directions.

[John Paul West]

Like Dave, I am new to this debate, but I don’t find your WM objection convincing. Suppose I see WM and then walk away. The next day, I want to think about WM. There are multiple ways I could go about this. I could intend the x I saw yesterday and then judge it likely to still exist as it was yesterday. Or I could intend the x which fills the location I was observing yesterday and then judge it likely that WM was not moved since yesterday. In both cases, I am intending something x really external to my mind (the WM as it was yesterday or the location of WM yesterday) and then judging that x is unlikely to have changed since yesterday, from which two statements I draw the syllogistic (though merely probable) conclusion “x presently exists”. Thus in each case I am intending something really external, but about which I may have made an error because of the merely probable judgement that it still exists. It seems to me we shouldn’t be surprised about such error, for even when the object is sensoraly present, our intention cannot “wrap itself around” the whole object.
As I said though, I am new to this debate, so perhaps I have made an obvious mistake?

[BV]

Thanks for the comment, David Brightly. >>The obvious answer to me is to retain 'intentional' and discard 'object'. We keep the directedness but drop the idea that it must terminate on a thing of some sort.<<

That makes sense. Intentionality would be like a ray in Euclidean geometry that proceeds from a point of origin in a straight line infinitely without every reaching a terminus. Perhaps you could say that the directedness is a property of the mental event rather than an object to which it is directed. There are theories like this. But how would that help us with the perceptual cases? They too are instances of intentionality. Suppose I travel to Washington and have a look at the WM. The directedness now terminates. More later perhaps.

[oz]

BV
> Perhaps you could say that the directedness is a property of the mental event rather than an object to which it is directed.

So instead of saying "The fundamental feature of every mental state is that it is directed beyond itself to an object", we say "The fundamental feature of every mental state is that it is directed".

The problem is that if the mental state involves a singular thought (e.g. the thought that Jake is thinking of Churchill), we cannot identify the thought without using the name which tells us which object the thought is about.

I think you accept that.

[oz]

Brightly
> The 'objectuality' must be put back in. I think this extra factor must be some degree of personal acquaintance with the WM

I can truly think about Zeus or say something about Zeus, without any acquaintance with Zeus.

Indeed I have just said something about Zeus.

[BV]

OZ at 2:46. Yes. there is a well-known problem with an internalist theory such as that of Husserl's when it comes to singular thoughts. But I stand by this:

D. E. Buckner speaks of an “. . . illusion that has captured the imagination of philosophers for at least a hundred years: intentionality, sometimes called object-dependence, a supposed unmediated relationship between thought and reality . . . .” (Reference and Identity in Jewish, Christian, and Muslim Scriptures: The Same God? Rowman and Littlefield, 2020, p. 195) Apart from his eliminativism about intentionality, Buckner is doubly mistaken in his characterization of it. No one except Buckner has, to my knowledge, characterized intentionality in general from Brentano on down as object-dependence, but it is standard, especially among analytic philosophers, to characterize it in terms of object-directedness.

I now add: that is a gross misrepresentation of intentionality theory from Brentano on. My point is doxographical, not philosophical. You got the history wrong. The main man in intentionality theory, Husserl, is an internalist, not an externalist -- which is not to say that his theory is adequate in the end.

[oz]

>My point is doxographical, not philosophical. You got the history wrong.

Nope. Part II to follow.

[BV]

John Paul West,

Thank you for your comment. I have read your comment three times, but I don't see that you are engaging my question.

[oz]

“We should also note the following. If the intentional nexus is not a relation (because some IOs exist and some do not), and the act-object nexus is non-contingent such that, necessarily, every act has an intentional object, then in the cases where the IO does not exist, and Meinongianism is false, the IO must be an immanent object.”

If Meinongianism is false, how can the act have an object that does not exist? For then some object, namely the object that the act has, does not exist.You should say that the act doesn’t have an object at all. But then “every act has an intentional object” is false, where e.g. the act is ‘thinking of Zeus’.

[BV]

>>If Meinongianism is false, how can the act have an object that does not exist?<< It does not exist with esse reale but with esse intentionale. This suggestion has been made. I am not endorsing it.

[oz]

>This suggestion has been made. I am not endorsing it.

But it seems to me that you endorse it in your rejection of P1 above. But the logic of your argument is not clear to me.

More later.