Standard argument against free will

The standard argument against free will is an argument that there exists a conflict between the possibility of free will and the postulates of determinism and indeterminism. A number of authors use this terminology.

A formal statement of the 'standard argument' can be phrased as follows:

1. The concept of determinism contradicts that of free will.

2. The concept of indeterminism also contradicts free will.

3. Some occurrences are governed by determinism, and all the rest by indeterminism.

all of which lead to the conclusion:

4. Free will does not govern any occurrences (does not exist).

The first two premises sometimes are referred to as the "deterministic" and "indeterministic" horns of the dilemma of determinism, the conflict between the intuition of personal autonomy and the intuition that the world proceeds in a manner independent of human concerns.

The formulation of the 'standard argument' above is abstract, avoiding particular definitions of 'determinism' and 'indeterminism' and 'free will' as matters best left out of the formulation itself. These definitions can then be subjected to further detailed discussion comparing the many views about what these terms mean.

The standard argument also commonly is phrased in terms of moral responsibility, rather than free will, as is mentioned later. This formulation restricts consideration to a subset of decisions that have moral implications. As such this approach is complicated by the need to discuss the connection between 'free will' and 'moral responsibility', which is itself seen from many perspectives.

An alternative wording of one premise

The third premise ('Some occurrences are governed by determinism, and all the rest by indeterminism') is sometimes stated as: "Either causal determinism is true, or it is false". This simple wording is defective. First, the use of 'true' and 'false' to characterize the claim that causal determinism and its negation are mutually exclusive and exhaustive of all possibilities is ambiguous: it is ambiguous regarding whether 'true' and 'false' are meant in a logical sense, part of a debate over definitions and semantics, or whether 'true' and 'false' are intended as an empirical claim about how things 'really' happen, a speculation based upon observations and evidence but extrapolating beyond any possible real-world attempt at verification. Second, supposing an empirical meaning is intended, it is unclear whether this sorting of events into two categories ('causal' or 'noncausal') is a blanket statement suggesting only one of these two categories is populated (an implausible situation), or intends to say that both categories are populated, and some events fall in one category, some in the other, which is the more plausible wording used in the third premise as it is stated above.

A final reservation is that 'causal determinism' is usually meant to be more restrictive than simply 'determinism', which last can be viewed many ways.

Moral responsibility

The idea of free will is closely connected to that of moral responsibility, inasmuch as there are reservations over holding an agent responsible for actions where they lacked free will to choose or, perhaps, lacked the capacity to exercise their decisions. Because our intuition of moral responsibility is strong and is, in fact, embedded in many countries' laws meting out punishment for crimes, the issue of moral responsibility is at the same time more personally challenging and more practically important than the 'standard argument', which may appear to be a largely academic dispute over the domains of applicability of determinism and indeterminism, or over terminology and usage.

Versions of the 'standard argument'

At this point, several different formulations of the 'standard argument' are presented. They all fit into the formulation of the introduction by adopting various definitions for the terms 'determinism' and 'indeterminism' and 'free will'.

Doyle

Doyle poses the argument as follows:

1. If our actions are determined, the will is not free.

2. If indeterminism and real chance exist, our will would not be under our control.

3. We could not be responsible for our actions if they are determined, or if they are random.

Doyle calls the first statement the 'determinism objection' and the second the 'randomness objection'. These two statements combine empirical and logical elements, and so complicate their discussion. Doyle does not include explicitly in the standard argument the premise that the choices of determinism and chance exhaust the possibilities. He continues (p. 17): "the "logical" standard argument against free will has been used by some philosophers ...to deny the existence of moral responsibility."

Fischer

Fischer frames the matter so 'free will', 'determinism' and 'indeterminism' in the introduction are replaced by a particular wording, and 'free will' is tied to a particular formulation of 'moral responsibility'. Fischer provides two versions. The first he attributes to William James, although it is Fischer's distillation of James' views, and is not stated this way by James himself:

1. Causal determinism means I have to act as I do, and thus am not morally responsible for my actions.

2. The only meaningful alternative to causal determinism is that how I act is a matter of luck, and thus I am not morally responsible for my actions.

and one claim, possibly of putative fact, or possibly merely a tautological proposition:

3. Either causal determinism is true, or it is false.

which statements in combination lead to the conclusion, which again may be a claim of fact, or a claim of logical consequence:

4. Therefore, I am not morally responsible for my actions.

His second formulation is as follows:

1. Causal determinism means I cannot do otherwise, and thus am not morally responsible for my actions.

2. If causal determinism is false (in a relevant way, that is, in the sequences leading to my behavior) then my actions are not appropriately connected to my prior states (that is, "my actions" are not in a genuine sense my actions) and thus I am not morally responsible for my actions.

and one claim, possibly of putative fact, or possibly merely a tautological proposition:

3. Either causal determinism is true, or it is false.

which statements in combination lead to the conclusion, which again may be a claim of fact, or a claim of logical consequence:

4. Therefore, I am not morally responsible for my actions.

If we look at Fisher's formulations, each premise involves multiple factors. For example, the first premise, Fischer's 'deterministic' horn of the 'dilemma', begins with a particular definition of 'determinism' connecting this concept to an agent's ability to act (free will + capacity to implement), and then connecting that issue with moral responsibility. The relationships between these three facets (determinism, personal autonomy, and moral responsibility) all are complex and debatable, and including them all in one premise complicates its acceptance, involving its discussion in ancillary issues. There is also an ambiguity as to whether the words 'true' and 'false' are meant in the same sense throughout, or appear sometimes in a logical sense (a matter of consistency in usage) and sometimes in an empirical sense (which is a matter of making assertions about the universe). A decision between these two interpretations decides whether the entire formulation is a purely theoretical logical construction without implication for the real world, or is intended to make an empirical claim about the real world. Further examination (see below) of Fischer's text will clarify that Fischer views the argument as having empirical consequences.

Rowlands

Rowlands also presents a syllogism with three premises and a conclusion, and uses this syllogism to identify the so-called dilemma of determinism:

 P1:  If our actions, choices and decisions are caused, they are not free

 P2:  If our actions, choices and decisions are not caused, then they still are not free

 P3:  Our actions, choices and decisions are either caused or they are not caused

  C:  Either way, our actions, choices and decisions are not free

This syllogism can be seen to set up a 'dilemma' for the select group of those who (i) want to believe in free will, and also (ii) accept the third premise of the syllogism. This group finds itself "in a spot of bother. The trouble takes the form of what is known as a dilemma...You're damned if you do and you're damned if you don't." Specifically, this group is 'damned' if they accept the first premise and 'damned' if they accept the second. In this formulation, the forced alternatives one must choose between are what are referred to as the 'horns of the dilemma', the 'horns' being the first two of the three postulates of the standard argument against free will.

Strawson

Strawson refers to a more compressed formulation as The Basic Argument:

 1.  Nothing can be causa sui — nothing can be the cause of itself.

 2.  In order to be truly morally responsible for one's actions one would have to be causa sui, at least in in certain crucial mental respects.

 3.  Therefore, no-one can be truly morally responsible.

Strawson's first premise avoids the term 'free will' by introducing the term causa sui, which carries with it some historical baggage about the understanding of 'causality'. This premise is the extension of 'causality' to every 'important' occurrence, a premise that appears in all formulations. His second premise replaces the two premises forming the customary 'dilemma' and introduces the rubric 'certain crucial mental respects'. The role of chance occurrences is not introduced directly. Strawson's formulation, like many others, involves 'moral responsibility', which complicates further discussion by introducing questions about what exactly is 'moral responsibility' and how is it connected to whatever one chooses to call causa sui. Another issue is the specifics of how causa sui is to be connected to 'free will'.

Objections to the 'standard argument'

The counter-intuitive conclusion of the 'standard argument', the conclusion that free will does not exist, is inescapable if one accepts the premises of the argument. So attempts to restore confidence in our intuition (and laws) based upon free will take two forms: (i) defining the terms 'determinism', 'indeterminism' and 'free will' in various ways that make the conclusion less clearly in opposition to intuition, and (ii) denying the third premise that it is an empirical fact that an 'either/or' formulation actually applies to the universe as we know it. Writers like Fischer take the first path, and others like Hume, Kant, and James take the second.

Hume

In 1739, David Hume in his A Treatise of Human Nature directed his attention to the conflict between determinism and moral responsibility (which involves free will). The notion is that one must examine more closely the relation between things regularly succeeding one another (descriptions of regularity in nature) and things that result in other things (things that cause or necessitate other things). According to Hume, 'causation' is on weak grounds: "Once we realize that “A must bring about B” is tantamount merely to “Due to their constant conjunction, we are psychologically certain that B will follow A”, then we are left with a very weak notion of necessity."

Russell discusses Hume's approach using the figure below:

(A) Chance         (B) Humean necessity    (C) Metaphysical necessity

 No regular succession   Regular succession    "Compelling" powers in objects

   ↑           ↑            ↑

              Moral realm?

      Humean necessity and the dilemma of determinism

According to Russell: "Hume is arguing that a middle path may be traveled between, on the one hand, a confused and unintelligible conception of necessity and, on the other, an erroneous belief in the existence of chance." Russell describes Hume's approach in terms of a dilemma, although Hume did not use that terminology: "In the light of the above diagram, it appears evident that Hume's strategy is to reveal that the dilemma of determinism, presented as an alternative between horns A and C, is a false dilemma. The standard dilemma of determinism is taken to be the choice between A and C and, according to Hume, it is a false dilemma because there is the third choice B." This statement of Russell's can be misleading, as Hume has not suggested that the 'dilemma' is replaced by three choices, instead of two, but rather, that moral decisions involve Hume's conception of causality.

According to Russell, Hume's position is that we do not attribute agency to the inanimate world. Further, our notion of cause and effect is a creation of our own minds based upon observation of sequences of events in the physical world. It is nothing but "an association of ideas which have been generated in the mind of the observer". But, says Russell, Hume concludes that our notion of personal autonomy is in the same boat. "We still have no reason to believe that such agents produce or bring about their actions, although our subjective experience may make us feel that they do." Regarding decisions with a moral aspect, according to Russell, Hume adopts the view that moral judgments are determined by character, which connects individuals to society, and which can decide an individual's actions.

"Where [actions] proceed not from some cause in the characters and dispositions of the persons who perform them, they...can neither redound to his honor, if good, nor infamy, if evil...the person is not responsible for the [action]...as it proceed from nothing in him that is durable or constant."

So at least some decisions are removed from the realm of simple events to become a matter of the quality of personal character, a trait not a whim of each instance, but shaped over time by interaction with society. "A Humean analysis of responsibility will investigate how these emotions [shame or guilt] lead us to be responsive to one another, in ways that support moral conduct and provide social penalties for immoral conduct. That is, its emphasis is less on people’s evaluation of themselves and more on how people judge and influence one another."

Kant

An influential approach to denying the empirical premise of the 'standard argument' was taken in the 1780's by Immanuel Kant, who had high confidence in the authority of intuition, and suggested that moral matters were to be analyzed as lying outside the rules governing material objects. "There is a sharp difference between moral judgments and judgments of fact...Moral judgments ... must be a priori judgments." Evidently, the 'standard argument' does not apply under these conditions, as our moral decision processes lie outside the reach of everyday causality.

According to Suppes:

"Unfortunately, not many philosophers..would be prepared today to follow Kant's way out of the dilemma of determinism."

However, contradicting that view, according to Velasquez:

"Many writers today agree with Kant. The philosopher/psychologist Steven Pinker, for example writes the following:

'Science and morality are separate spheres of reasoning...[more of this quote is provided]'

Here Pinker is agreeing with Kant. ... So are we free or determined? Are we responsible agents or passive victims? Was Darrow right? Or was Sartre right? Or were both right as Kant and Pinker suggest?"

James

In 1884 William James presented a talk entitled The Dilemma of Determinism in which he attacked the notion of determinism as simply a convention of the mind used to organize our inchoate sense impressions:

"I myself believe that all the magnificent achievements of mathematical and physical science — our doctrines of evolution, of uniformity of law, and the rest — proceed from our indomitable desire to cast the world into a more rational shape in our minds than the shape into which it is thrown there by the crude order of our experience. The world has shown itself, to a great extent, plastic to this demand of ours for rationality. How much further it will show itself plastic no one can say...If a certain formula for expressing the nature of the world violates my moral demand, I shall feel as free to to throw it overboard, or at least to doubt it, as if it disappointed my demand of uniformity of sequence...The principle of causality, for example, — what is it but a postulate, an empty name covering a demand that the sequence of events...manifest a deeper kind of belonging of one thing with another than the mere arbitrary juxtaposition which now phenomenally appears?"

Having, he thought, demolished the notion of determinism as a mere device of the mind, James then proposed that chance was the more realistic alternative:

"And this at last brings us within sight of our subject. We have seen what determinism means: we have seen that indeterminism is rightly described as meaning chance; and we have seen that chance, the very name of which we agreed to shrink from...means only the negative fact that no part of the world, however big, can claim to control absolutely the destinies of the whole"

James' definition of chance is very much broader than that of many modern writers, who view chance as just a roll of the dice. In particular, James' view is compatible with Kant's, although not the same.

Doyle

Doyle's formulation of the 'standard argument' is presented above. He has detailed a number of objections. One of these is the following:

"Physical determinism is not 'true' because physics is empirical, not logical. And the empirical evidence has never justified the assumption of strict determinism."

This argument is one of those made by Fischer as well, among those discussed next.

Fischer

Fischer's formulation of the 'standard argument' is presented above. He proceeds as follows:

"I shall argue that neither the [first] premises of the parallel arguments nor the [second] premises are true, and thus the argument is unsound for two separate reasons. I shall further argue that similar considerations help to establish the failure of both the deterministic and indeterministic horns of the dilemma." [Numbering has been changed to match the presentation of this article.]

He then argues that the view that causal determinism implies that we never have the freedom to do other than we actually do, for example, the 'consequence argument' of Van Inwagen, fails because it exaggerates beyond any possibility of real-world attempt at verification, and presupposes a form for scientific laws beyond any supportable claim inasmuch as these laws are limited in scope and their precise form is restricted to what we know of them today.

"If we knew that determinism were true and we also knew both the natural laws and the complete description of the universe at the present (or at any point in the past), we could predict with certainty whether or not the house will be destroyed by an earthquake. But in fact we do not know the natural laws or whether they are deterministic; and we do not have available such a description."

Actually, this argument critiques not the definitions of various terms (impacting premises 1 and 2 of the introduction), but the empirical premise of the standard argument (item 3 in the introduction) and is close to what Hume, Kant and James also claim.

However, Fischer views this criticism to be controversial, and proposes instead "to attack the contention that moral responsibility requires freedom to choose and do otherwise". Fischer "contend[s] that it is not the case that all causally deterministic sequences crowd out moral responsibility". "We have found plausible and 'independent' arguments for the resilience of moral responsibility to the truth or falsity of causal determinism--arguments that are not simply statements of the desirability of such resilience. In the end, then, moral responsibility is neither inconceivable nor puzzling."

Russell

Russell's analysis of Hume is summarized above. His purpose in discussing Hume is in part to illuminate more modern concerns. In particular, he notes the oddity that, although we want to believe in our autonomy and independence from natural law, yet we could hardly be held responsible if the world itself was completely unpredictable, so that we had no idea of how to act, because we had no idea what would result. So we have a question: Can we ourselves intervene to disturb the sequence of nature? Russell speaks of 'weak' and 'strong' interpretations of causality, weak enough to allow personal autonomy, but strong enough to ground decisions in reasonable expectation of their consequences. Do we have here a question of the interpretation of causation (that is, a semantic issue) or is it an empirical matter? And if the latter, exactly what rules govern our empirical exploration, and why can't they be applied to ourselves? Russell quotes Spinoza:

"Most of those who have written about the emotions and human conduct seem to be dealing not with natural phenomena that follow the common law of Nature but with phenomena outside Nature. They appear to go so far as to conceive man in Nature as a kingdom within a kingdom. They believe that he disturbs rather than follows Nature's order."

It should be noted that Spinoza felt this situation should be corrected. His view was: "God is not the transcendent creator of the universe who rules it via providence, but Nature itself, understood as an infinite, necessary, and fully deterministic system of which humans are a part."

Remarks

The 'empirical premise' of the standard argument, that all events are subject to physical laws as we know them, is an extrapolation of what is known of the 'laws of nature' beyond any foreseeable ability for verification via observation. The resolution of these matters falls within the subjective-objective dichotomy.