Friday, May 25, 2007

A Quick Look at a Simple Definition of LFW

(This post may get deleted, in order to harmonize the discussion on free will.)

Godismyjudge has provided a new simple definition of LFW:

An agent has free will if and only if the agent is able to do otherwise than what he will do.

This definition is not harmonious.

Let's break it down.

1) An agent has (present tense) free will This line is understandable. It is an assertion about a present state of the agent.

2) if and only if This line is also understandable. A definition is about to follow.

3) the agent is able (present tense) to do otherwise This line is mostly understandable. It is a statement about a present ability to do something, where the something is defined negatively, and we are waiting for that something. So we continue to the last line

4) than what he will do (future tense) This line by itself is understandable. What is being described is a future action of the agent.

The lack of harmony appears when we combine 3 and 4 in view of 1. That is to say, when we speak of a present quality that depends on a comparison between the present and the future.

This lack of harmony is evidenced several ways:

First: the lack of harmony is evidenced by the fact that in order for the statement to have meaning, there must be (present tense) something that the agent will do (future tense). That would suggest that the future already exists, in some sense. Such a suggestion unravels the entire sweater of LFW, as has previously been seen.

Second: the lack of harmony is evidenced by the fact that my being able to eat steak today, and my actually not eating steak tomorrow would fully qualify under the definition. In other words, if I eat steak on Friday, and not on Saturday, I have proven that I had free will on Friday, under this definition. This would suggest that a completely deterministic (even a mechanically deterministic) world would include agents with free will, because they had the ability to do otherwise than what they did do.

I call these a lack of harmony, not because the definition itself is incoherent, but because these results do not harmonize with a non-compatibilist mentality.

In other words, if the simple definition above defines LFW, then Calvinists all agree that man has LFW, because we all agree that we can do something different today than we will do tomorrow.

-Turretinfan

3 comments:

Anonymous said...

Dear Turretinfan,

You found two forms of “lack of harmony” with my simple definition of libertarian freewill.

LFW: “An agent has free will if and only if the agent is able to do otherwise than what he will do.”

Here’s the first:

TF: First: the lack of harmony is evidenced by the fact that in order for the statement to have meaning, there must be (present tense) something that the agent will do (future tense). That would suggest that the future already exists, in some sense. Such a suggestion unravels the entire sweater of LFW, as has previously been seen.

This is of course not a contradiction found within the definition of LFW, but rather with other views held by those that hold to LFW. Accordingly, even if your argument is granted, the definition is not shown to be incoherent, as you note.

Your argument is less than clear, but here’s what I made of it. (If you meant something else, please reformulate.)

P1: necessarily, if LFW is correct, there is something that the agent will do
P2: but there being something we will do entails that the future exists in some sense
C1: therefore if LFW is correct, the future exists in some sense
P3: TF has shown that if the future exists in some sense, LFW does not exist
C2: therefore LFW lacks harmony

My answer is to distinguish P2 (and also P3), by defining the sense the future exists today. As we had discussed, if the future is defined as

1: the set of true future tense propositions,

the future exists today. If on the other hand the future is defined as

1’: the events that will happen,

the future does not exist today.

I had previously argued (at least to my satisfaction :-)) that 1’ is incompatible with LFW but 1 is not. So no lack of harmony is shown, unless you can demonstrate that LFW requires 1’, which your argument does not.

Here’s your second argument.

Second: the lack of harmony is evidenced by the fact that my being able to eat steak today, and my actually not eating steak tomorrow would fully qualify under the definition. In other words, if I eat steak on Friday, and not on Saturday, I have proven that I had free will on Friday, under this definition. This would suggest that a completely deterministic (even a mechanically deterministic) world would include agents with free will, because they had the ability to do otherwise than what they did do.

Your argument could be construed in two ways. Let’s define T as today and T+1 = sometime later today and T+2 = tomorrow.

The major premise of your argument is either:

2: At T, I had the ability to eat steak at T+1, and actually did eat steak at T+1, but did not at T+2.

2’: Or at T, I had both the ability to eat steak at T and actually eat steak at T, but did not at T+2.

My response is that neither 2 or 2’ meets the definition of LFW.

2 has two future points in which you will either eat or not eat steak. It would be a contradiction if you will and will not eat steak at T+1. It would also be a contradiction if you will and will not eat steak at T+2. But it would not be a contradiction if you will eat steak at T+1 and will not eat steak at T+2. In this sense (an alternative in a self contradictory pair), eating steak is at T+1 is not “otherwise” than eating steak at T+2. If one part of the pair is at T+1 and the other is at T+2, then only in a divided sense are they alternatives or “otherwise”.

Further, at T, You will eat steak is true (as you will eat steak at T+1). Hence, eating steak at T+1 is not “otherwise than what you will do”.

2’ also has the “otherwise” being at two different points in time. But I am more sympathetic to 2’, because there are present and future, not future and future. 2’ avoided the issue of not being otherwise, because you will eat steak. Indeed, you will not eat steak (at least as far as is given). So at T you are doing otherwise than what you will do.

However, 2’ entails:

3: an agent may have both ability concerning an event at T and use that ability at T.

In some sense 3 is plausible. But not in this sense:

4: an agent is able to be a cause, such that eating steak is the effect

Because causation works foreword in time (else how shall we measure time?). So the ability is at T and use is at T+1. I believe physical ability falls under a subset of causation.

Nor in this sense is 3 the case:

5: an agent’s ability relates to his potential or the mere possibility of him doing something, his use of that ability relates to something actual

Because what is actual is not potential or merely possible.

I would suggest that 4 & 5 are the most likely senses in which LFW is to be taken. Therefore, 2’ (or 3) is an unlikely candidate for validating LFW.

However, I admit that we are likely to say at T you have the ability to eat steak. But by this we mean that:

6: in a 4 or 5 sense you had that ability at T-1, and at T you have the ability to continue eating steak at T+1.

6 however, does not permit 2’ to be an example of LFW.

It seems that if you are eating steak at T, you are no longer able to eat steak at T. The potential is now actual, the causal ability you had, has been laid down.

It’s not apparently clear to me if there is some alternative sense to ability, besides 4 & 5, one in which use of ability is not temporally subsequent to ability. At least a sense that is not equivalent to 6. If there is, please provide it. But as things stand, I would be hesitant to accept 3.

So 2’ does not demonstrate a lack of harmony within LFW or with the view held by those who hold to LFW.

God be with you,
Dan

Turretinfan said...

Dear Godismyjudge,

In your response you posited two definitions of the future:

1: the set of true future tense propositions, and

1': the events that will happen.

You seem to think that definition 1 has some kind of legitimacy. I don't agree.

True future tense propositions describe the future, they are not the future, just as Genesis 1 is not itself the Creation of the World, but simply describes it. This seems blindingly obvious to me, and I know you are no fool, so I'm wondering why this hasn't occurred to you?

Of course, if 1 is out (as it surely must be), and (in your view) only 1' remains, then by process of elimination 1' is all that is left, and you have acknowledged the unravelling effect that 1' has on your system of thought.

So, please explain how the future IS actually a set of propositions as opposed to the events the propositions describe.


You provided a lengthy response to the second source of disharmony, perhaps from a mistaken understanding of what I was trying to convey. I note that you're attempt to provide my major premise seems to be way off of what I intended. Let me see if I can place it in syllogistic form for you:

P1: A harmonious definition of LFW is such IFF it avoids acceptance by hard determinists AND compatiblists.
P2: I have provided a way in which your definition would be acceptable to hard determinists and compatiblists.
C: Therefore your definition is not a harmonious definition.

Do you follow the syllogism?

Perhaps you dispute P2.

But recall the definition:

"An agent has free will if and only if the agent is able to do otherwise than what he will do."

The example meets the definition because:

Sub-premise-2-1 (S1): I am an agent.
S2: If I meeting the part following the IFF, then I have free will in the definition.
S3: If I eat steak today it is proof that I am able to eat steak.
S4: Not eating steak is an activity that is otherwise than eating steak, and vice versa.
S5: Not eating steak is what I will do.
S6: Combining S4 and S5 eating steak is otherwise than what I will do.
S7: Combining S6 and S3 If I eat steak today it is proof that I am able to do otherwise than what I will do.
S8: I eat steak today.
Sub-conclusion-2-1(SC1): I am able to do otherwise than what I will do.

Do you follow? At what point do you disagree?

-Turretinfan

Turretinfan said...

Dear Godismyjudge,

You state: "Because causation works foreword in time (else how shall we measure time?)"

While it is far from the main thrust of your response, I thought I'd point out the answer. We measure time by simultaneous correlation, not by succession. We measure time by correlation to the position of the sun and shape of the moon.

-Turretinfan