[7.] Peirce’s Modal Shift

 

As we have seen, Peirce readily uses the modal notions of possibility and necessity in his evolutionary cosmology. In particular, his tychism is that idea that there is real “absolute chance” in the form of actual events that are possible but not necessary, i.e., that are contingent.

 

But we have not yet seen how he defined the modal terms “possibly” and “necessarily.” In these lectures, we will examine how Peirce defined the modal terms during this time (1891-1893).

 

We will also see how, in 1896, Peirce changed his view of how modal terms should be defined. I call this change Peirce’s “modal shift.” [1]

 

This change has effects in other parts of his philosophical system. In particular, it results in revisions of his

·         scholastic realism,

·         pragmatism, and

·         synechism.

 

What’s more, as a result of this change of view on modality, these doctrines come to be more tightly connected with one another than they had been before.

 

In this set of lectures, we will examine his the modal shift itself. In later lectures, we will examine the effects this change had in his philosophical system, in particular, for his:

·         scholastic realism,

·         pragmatism, and

·         synechism.

 

 

[7.1.] Different Types of Modality.

 

To explain Peirce’s first view of modality, I will focus primarily on the concept of possibility, but I will also on occasion refer to necessity as well. (Keep in mind that “necessity” can be defined in terms of possibility: “It is necessary that p = It is not possible that it is not the case that p.” This is not a deeply helpful definition, however, unless some more substantial definition of “possible” can also be provided.)

 

Traditionally, philosophers have distinguished different types of possibility, including...

·         logical possibility, which has to do with what can be true given the laws of logic; the logically possible is that which does not involve a contradiction (e.g., it is logically possible that George W. Bush is married, and it is also logically possible that George W. Bush is a bachelor, even though in fact he is not a bachelor), while the logically impossible does involve a contradiction (e.g., it is not logically possible that there is a married bachelor, or that there is a square with five sides).

·         physical possibility, which has to do with what can be true given the laws of physics (e.g., it is physically possible for jet airplanes to fly, but it is not physically possible for any man-made object to travel faster than the speed of light).

·         practical possibility, which has to do with what can be true given the practical limitations on human abilities (e.g., it is practically possible for humans to build very tall skyscrapers, but it is not practically possible for us to travel to another solar system within a human lifespan).

·         epistemic possibility, which has to do with what might be true given our current knowledge of the actual world (e.g., it is possible that Dr. Alberg shaved this week, meaning that I don’t know whether or not he did).

 

 

[7.1.] The Information-Relative Account of Modality. [2]

 

During the period of his cosmological papers, Peirce believed that every sort of possibility could be defined in terms of states of knowledge or information: “... the Possible, in its primary meaning, is that which may be true for aught we know, that whose falsity we do not know.” (3.374, 1885; “On the Algebra of Logic”)[3]

 

The modal expression...	should be understood to mean...
“It is possible that p.” (“It may be the case that p”; “It might be the case that p”)   E.g., you ask me whether it is going to rain today. I say “Possibly, it will rain.”	“It is not known to be false that p.”     This should be taken to mean that, given my present state of information (including everything I know about the weather), I do not know that it won’t rain.
“It is necessary that p.” (“It must be the case that p”; “It has to be the case that p”)   E.g., I know that my keys are either in my office or in my car. I check my car and discover that they are not there. So I say “My keys must be in my office” (“It must be the case that my keys are in my office.)”	“It is known to be true that p.”     This should be taken to mean that, given my present state of information (including everything I know about my keys), I do know that they are in my office.

 

These are straightforward examples of epistemic possibility. They all refer to actual states of information, the state of knowledge that an actual person might find himself or herself in.

 

But Peirce took, not just epistemic possibility, but all types of possibility, to be definable in terms of states of information or knowledge.

 

However, he did not think other types of possibility could be defined in terms of actual states of knowledge in which actual human beings might find themselves.

 

Other sorts of modality must be defined in terms of hypothetical states of knowledge:

 

To say that a fact is possible means primarily that it is not known not to be true, that our knowledge leaves room for it. But there are, besides, kinds of possibility, which are determined not by the knowledge which we happen just now to have, but by every conceivable state of information. (W 5:330, 1886; “Qualitative Logic”)

 

For example, physical possibility and necessity are defined in terms of the state of information of a hypothetical person, one who is “thoroughly acquainted with all the laws of nature and their consequences, but … ignorant of all particular facts” (CP 4.66, 1893 [handout, selection 1]):

·         “It is physically possible that p” = such a hypothetical person would not know that it is false that p.

·         “It is physically necessary that p” = such a hypothetical person would know that it is true that p.

 

And practical possibility and necessity are defined in terms of the state of information of a hypothetical person, one who “know[s] what the resources of men are, but not what their dispositions and desires are” (ibid.)

 

Peirce defines logical possibility and necessity in a similar way. But he distinguishes two sorts of logical possibility and necessity, or, in his words, types of modality “which are of special interest to the logician more than to other men.” (CP 4.67, 1893 [handout, selection 1]:

 

essential possibility and necessity: involves “the states of information in which we are supposed to know nothing, except the meanings of words, and their consequences ... That is essentially or logically possible which a person who knows no facts, though perfectly au fait at reasoning and well-acquainted with the words involved, is unable to pronounce untrue. The essentially or logically necessary is that which such a person knows is true.” (ibid.)

 

substantial possibility and necessity: involves “the states of information ... in which we are supposed to know everything ... the substantially possible refers to the information of a person who knows everything now existing, whether particular fact or law, together with all their consequences. This does not go so far as the omniscience of God; for those who admit Free-Will suppose that God has a direct intuitive knowledge of future events even though there be nothing in the present to determine them. ... The terms, substantial necessity and substantial possibility, however, refer to supposed information of the present in the present, including among the objects known all existing laws as well as special facts.” (ibid.)

 

Substantial modality is the sort of modality required by tychism. We can understand that doctrine as claiming that there are future events that are substantially contingent, that is, substantially possible but not necessary.

 

(Peirce defines mathematical and metaphysical modalities similarity, in terms of the states of information of “the most perfect mathematician or metaphysician.” (ibid.))

 

So, early on Peirce took all senses of the modal terms to be definable in terms of states of information.

 

 

[7.3.] Two Types of Modal Realism.

 

At first, it might seem that there is a conflict between, on one hand, Peirce’s information-relative account of modality, and on the other, his realism about modality.

 

Peirce clearly believes that there is real (substantial) contingency in the world: this is his doctrine of tychism.

 

But as we have seen, he defines “real” as that which is independent of what anyone in particular thinks about it.

 

How, then, could Peirce consistently maintain that some events are really contingent while at the same time defining “contingent” (and other modal terms) in terms of states of information?

 

Peirce anticipated this objection and replied as follows:

 

To conclude from the above definitions [i.e., the IR account] that there is nothing analogous to possibility and necessity in the real world, but that these modes appertain only to the particular limited information which we possess, would be even less defensible than to draw precisely the opposite conclusion from the same premises. It is a style of reasoning most absurd. (CP 4.68, 1893; modality hand-out, selection 1)

 

Clearly, Peirce thought that there was no conflict between the IR account and his realism about contingency.[4]

 

We can clarify Peirce’s position here by distinguishing two sorts of question:

(1)    What modalities are exemplified? Assuming that every actual event is possible, are there actual events that are contingent? Or, are all actual events necessary?

(2)    What is the metaphysical status of modality? Is it real (i.e., independent of what anyone thinks about it)? Can all modal terms be adequately defined in terms of states of information?[5]

 

The first sort of question asks whether there are any events that fall under concepts such as necessary and contingent, while the second sort of question asks about the nature of necessity, contingency and possibility themselves. Peirce’s tychism addresses a type (1) question about what sorts of modality are exemplified: it asserts that some future events are substantially contingent. On the other hand, the IR account addresses type (2) questions about the metaphysical status of modality.

 

Prior to the modal shift, Peirce held that all types of possibility, necessity and contingency ought to be understood in terms of (actual or hypothetical) states of information and that this was compatible with there being something “analogous” to modality “in the real world.” This suggests a range of responses to type (2) questions, with Peirce’s 1893 view occupying a moderate ground between two extremes:

 

(a)    strong modal realism: modality (possibility, necessity, contingency) is real, and at least some types of modality cannot be defined in terms of states of information, even hypothetical states of information.

(b)    weak modal realism: modality is real and all types of modality can be defined in terms of states of information.

(c)    modal anti-realism: modality is not real.

 

Prior to the modal shift, Peirce was a weak realist. He took modality, including future substantial contingency, to be real, yet he took the IR account to be sufficient to define all types of modality.

 

 

[7.4.] The Modal Shift.

 

The modal shift of 1896 was a move from weak to strong realism. What Peirce changed his mind about was not whether events can accurately be described as contingent, possible, necessary, etc. Instead, he changed his mind about whether one sort of modality, viz. substantial modality, could be defined in terms of states of information.

 

The modal shift occurred in the second of two articles in which Peirce reviewed Ernst Schröder’s Exact Logic (1895, 3vv.):

·         “The Regenerated Logic,” The Monist, 7, 1896, 19-40. (CP 3.425-55; selection 2 in the “modality” readings handout is an excerpt from this)

·         “The Logic of Relatives,” The Monist, 7, 1897, 161-217. (CP 3.456-552; selection 3 in “modality” readings handout is an excerpt from this)

 

Peirce moved from weak realism about modality to strong realism in the course of writing these articles.

 

In the first article, he once again asserted that all senses of “possibility” can be adequately defined in terms of states of information:

 

... possibility may be understood in many senses; but they may all be embraced under the definition that that is possible which, in a certain state of information, is not known to be false. By varying the state of information all the varieties of possibility are obtained. (3.442, 1896; modality handout, selection 2)

 

And he went on to repeat his information-relative definitions of the two types of logical possibility:

 

Thus, essential possibility is that which supposes nothing to be known except logical rules. Substantive possibility, on the other hand, supposes a state of omniscience. (3.442, 1896; modality handout, selection 2)[6]

 

But in the second article, “The Logic of Relatives,” he abandoned this view:

 

I formerly defined the possible as that which in a given state of information (real or feigned) we do not know not to be true. But this definition today seems to me only a twisted phrase which, by means of two negatives, conceals an anacoluthon. We know in advance of experience that certain things are not true, because we see they are impossible. (3.527, 1897; modality handout, selection 3)

 

But this change of heart was not a total rejection of the information-relative account. As it turned out, Peirce would continue to define some types of possibility in terms of states of information, including essential logical possibility:

 

As I remarked in the last Monist ["The Logic of Relatives," selection 3 in this document], my old definition of the possible as that which we do not know not to be true (in some state of information real or feigned) is an anacoluthon.  The possible is a positive universe, and the two negations happen to fit it, but that is all.  Of course, there is a general logical possible that is no more than I defined it.  But there is also a possible which [is] something else. (8.308, 1897; modality handout, selection 4)

 

What Peirce changed his mind about in 1896 was whether one type of possibility in particular could be defined in terms of states of information, viz. substantial possibility, the type of possibility required by tychism.

 

 

Stopping point for Wednesday October 17. For next time, read the rest of the “Peirce on Modality” handout (pp.4-10).