Sunday, November 22, 2015

On Certainty

Certainty is the state of being or feeling certain, and is freedom from doubt. If something is certain, then it is free from, or beyond, doubt.

To claim absolute certainty about the truth of a proposition may be to claim the right to ignore or be blind to further evidence affirming or denying the truth of that proposition.

Logically, if you say that you're certain about something, then you should be prepared to wager everything on it. If you're not prepared to wager everything on it, then you must not really be certain about it (unless there are psychological, social, or other factors that prevent you from making such a wager, despite your alleged certainty).

Of course, if you say that it's not in your nature to wage everything on something, then your reluctance to wager everything (or to wager anything at all) may be due to your recognizing that you can never be absolutely sure of, or certain about, anything.

Benjamin Franklin's well-known saying (now so often quoted that it has become a cliché) regarding certainty appeared in a letter to the French scientist Jean-Baptiste Leroy, written on November 13, 1789: "In this world nothing can be said to be certain, except death and taxes."1

Other certainties may include mathematical certainties (such as "2+2=4"), physical certainties (such as "marble is harder than limestone"), cosmological certainties (such as "Ganymede is the largest moon of Jupiter"), and historical certainties (such as "The Battle of Bosworth Field was fought in 1485").

F.W.J. Schelling (1800) says that "for certainty in theory we lose it in practice, and for certainty in practice we lose it in theory" ("über der theoretischen Gewissheit geht uns die praktische, über der praktischen die theoretische verloren").2

I can't be more certain of anything than I am of the fact that some (though perhaps not all) lost opportunities are lost permanently. Perhaps this is merely certainty of my own (and more generally of human) finitude. I may only be certain of my own fallibility and uncertainty.

Believable fiction may depend on fictionalization of the real and realization of the fictional.

If to be thoughtful is to be careful, and to be thoughtless is to be careless, then you might suppose that to think is to care. But unfortunately that's not the case.

Isn't the term "certain knowledge" redundant? Either we know something or we don't. If we're certain that we know something, then we may more precisely be said to have attained "certainty of knowledge" or "epistemic certainty."

It may always be possible to say "I don't know" with greater certainty than "I know."

If I'm feeling pain, and it's a pleasurable pain, then I may be uncertain whether I'm actually feeling pain.

Epistemic duty (or the ethics of knowledge) may require that we have adequate grounds for saying we're certain about something. It may also require that we have adequate grounds for saying we're doubtful about something.

Wittgenstein asks, "Doesn't one need grounds for doubt?"3 But it can be argued that we aren't required to believe a proposition to be true just because we can't prove it to be false.

If we qualify our statements about certainty by saying, "We think (but aren't sure) we're certain about that," then we may be saying we're relatively certain, but not absolutely certain. Relative certainty may be susceptible to doubt, while absolute certainty may be immune to doubt. The former may be much easier to attain than the latter.

If we tell others that we know something, then we may have a duty to let them know when we're not absolutely certain that we know. Our claim to know something requires some degree of certainty on our part. We don't always have to be absolutely certain about something in order to properly claim to know it. Relative certainty may be sufficient in some cases. But we must at least have met some threshold level of certainty about something in order to properly claim to know it.

The term "objective certainty" may be ambiguous, insofar as it may refer to either the objectivity of the fact that certainty has been attained or the objectivity of the reasons for which certainty has been attained. In the first case, certainty can objectively be said to have been attained, while in the second, certainty can be said to have been attained objectively. It may be important to distinguish between the two cases. For example, Bill can objectively be said to be certain about something without his having objective reasons for being certain about that thing.

Subjective certainty may include psychological and intuitive certainty. Objective certainty may include epistemic, propositional, and evidential certainty. 

One way of expressing subjective certainty may be to say, "I'm certain, but I could be wrong." This way of hedging one's bets about certainty may be a way of immunizing oneself from error—
or it may just be a way of equivocating.

Subjective certainty doesn't entail objective certainty. We may be subjectively certain about something, but objectively wrong about it.

There may be various kinds of certainty (psychological, moral, epistemic, deductive, and statistical). There may also be various degrees of certainty (which may be expressed by such adjectives as "absolute," "high," "low," or "relative.")

However, Peter Unger (1975) argues that there are no degrees of certainty, and that "certainty" is an absolute term, like "flat." Just as nothing (or hardly anything) in the world is (absolutely) flat, nothing (or hardly anything) in the world is (absolutely) certain. Unger thus argues for scepticism regarding the possibility of knowledge, by saying that since knowledge entails (absolute) certainty, nothing (or hardly anything) is known.4,5

Peter Klein (1981) also argues that knowledge entails absolute (psychological and evidential) certainty, and that if S knows that p on the basis of some evidence e, then e renders p absolutely certain for S. Klein argues that S knows that p only if p is absolutely certain for S, but he makes clear that he's not saying that the ordinary understanding of the meaning of the term "knowledge" restricts the term to only those cases in which a proposition p is absolutely certain.6 There may be relative as well as absolute uses of the term "certainty."

Harry Frankfurt (1962) says that there may indeed be degrees of certainty, and that some things may be regarded as more certain than others. We may be willing to wager more on the truth of one proposition than we're willing to wager on the truth of another proposition, if we regard the one's truth as more certain than the other's.7

Jason Stanley (2008) says that knowledge doesn't entail certainty, and as an example he provides the statement, "I know that Bill came to the party. In fact, I'm certain that he did," which seems to show that we can know something without being certain of it. As another example, he provides the statement, "John knows that Bush is a Republican, though, being a cautious fellow, he's only somewhat certain of it," which seems to show that we can know something without being completely certain of it. Stanley explains that there's nothing strange or odd about ascribing knowledge of something to someone who's not completely confident that she does indeed know that thing. We don't automatically object when someone is described as knowing something of which she isn't certain.8

Stanley also explains that fallibilism in epistemology (the theory that knowledge is compatible with a lack of certainty) is compatible with both "the knowledge norm for assertion" (that we should assert p only if we know that p) and "the certainty norm for assertion" (that we should assert p only if we're certain that p). Thus, while fallibilism holds that we can know p without being certain that p, it doesn't hold that we can say we know that p if we don't know or aren't certain that p.9

Gaining greater knowledge of something may sometimes lead us to feel less certain about that thing. By learning more about that thing, we may sometimes develop a greater appreciation of the limits of our knowledge. 

Second-order certainty may be exemplified by such statements as, "I feel certain that I'm certain."

The certainty of faith may be an example of psychological, moral, theological, or doctrinal certainty. It may be based on belief in truths that cannot be proved by reason or that surpass our understanding. It may be based not on empirical proof or scientific testing, but on the power of authority, scripture, testimony, or revelation.


1Benjamin Franklin, The Works of Benjamin Franklin, Vol. XII, Letters and Miscellaneous Writings, 1788-1790.
2F.W. J. Schelling, System of Transcendental Idealism, translated by Peter Heath (Charlottesville: University Press of Virginia,1978), p. 11.
3Ludwig Wittgenstein, On Certainty, edited by G.E.M. Anscombe and G.H. von Wright, translated by Denis Paul and G.E.M. Anscombe (Malden: Blackwell Publishing, 1969), p. 18c.
4Peter Klein, Certainty: A Refutation of Scepticism (Minneapolis: University of Minnesota Press, 1981), p. 131.
5Peter Unger, Ignorance: A Case for Scepticism (Oxford: Oxford University Press, 1975), p. 49.
6Klein, Certainty: A Refutation of Scepticism, p. 117.
7Harry G. Frankfurt, "Philosophical Certainty," in The Philosophical Review, Vol. 71, No. 3 (July 1962), pp. 303-327.
8Jason Stanley, "Knowledge and Certainty," in Philosophical Issues, Vol. 18, Issue I (September 2008), pp. 41-42.
9Ibid., p. 56.

Sunday, November 1, 2015

The Dispositional-Occurrent Distinction

What exactly is a disposition? Tim Crane (1996) offers the following preliminary definition: 

“a disposition is a property (such as solubility, fragility, elasticity) whose instantiation entails that the thing which has the property would change or bring about some change, under certain condtions.”1 

      Stephen Mumford (2011) offers another preliminary definition, by saying that the term “disposition” may refer to a type of property, state, or condition that, under certain circumstances, provides for the possibility of some further specific state or behavior.2 
      Thus, for example, if we say that “ice cubes have a disposition to melt at room temperature” (or simply that “ice cubes melt at room temperature”), we are saying that under those conditions (room temperature), the disposition of ice cubes to melt will be manifested.
      Jennifer McKitrick (2003) explains that some marks of dispositionality include the following: (1) a disposition has a characteristic manifestation, (2) a disposition is triggered in certain types of circumstances (it has certain circumstances of manifestation), (3) if something has a disposition, then certain subjunctive or counterfactual conditionals are typically true of it (e.g. if a fragile glass has a disposition to break when dropped, then the conditional “If it were dropped, then it would break” is true of it), and (4) once the manifestation and the circumstances of the manifestation have been identified, the disposition can be referred to by an overtly dispositional locution, such as “the disposition to so and so” (e.g. “the disposition to break when struck” refers to fragility, and “the disposition to dissolve in water” refers to water-solubility).3
      Dispositions may be ascribed to kinds (e.g. “paper is flammable”), objects (“that sheet of cardboard is flexible”), persons (“I tend to agree with you”), and conditions (“the weather’s likely to turn chilly today”).
      Crane (1996) explains that there has been much debate about the question of whether the ascription of a dispositional property to a thing entails that certain counterfactual conditionals are true of it. If this is indeed a criterion of dispositional properties, then, for example, saying that a glass is fragile entails the conditional: “if the glass were struck with sufficient force, then it would break.”4
      Another example of a conditional analysis of dispositions would be to say that the elasticity of a rubber band entails the conditional, “if the rubber band were pulled, then it would stretch, and if the pulling force were removed, then it would return to its original length.”
     However, C.B. Martin (1994)5 introduces the notion of a “finkish” disposition in order to show that the ascription of a dispositional property to a thing does not necessarily entail that an associated conditional is true of it. Thus, the testing of a “finkish” disposition may not result in the manifestation of the disposition, because the testing itself may cause the disposition to be lost. Alternatively, in a reverse finkish case, a disposition may be absent, but may be gained when it is tested for.6
      Sungho Choi (2011) explains that dispositional maskers have also been introduced as counterexamples to the notion that dispositional properties can be analyzed in terms of simple counterfactual conditionals. If a disposition is masked, or if there is an antidote to it, then it may not be manifested, even under the appropriate stimulus conditions. Choi argues, however, that the absence of dispositional maskers may be implicitly or explicitly implied by dispositional ascriptions, and that the failure of masked dispositions to be manifested does not necessarily threaten the simple conditional analysis of dispositions.
      Choi (2009) also explains that the two most widely debated versions of the conditional analysis of dispositions are the simple conditional analysis of dispositions (SCA) and the reformed conditional analysis of dispositions (RCA). The SCA may be expressed as follows:

      “Something x is disposed at time t to exhibit manifestation m in response to being situated in stimulating circumstance c iff, if x were to be situated in c at t, it would exhibit m.”7

The RCA, formulated by David Lewis (1997)8, may be expressed as follows:

      “Something x is disposed at time t to exhibit manifestation m in response to stimulus s iff, for some intrinsic property B that x has at t, for some time t’ after t, if x were to undergo stimulus s at time t and retain property B until t’, s and x’s having of B would jointly be an x-complete cause of x’s exhibiting manifestation m, where an x-complete cause is a cause complete in so far as havings of properties intrinsic to x are concerned, though perhaps omitting some events extrinsic to x.”9

Lewis’s RCA is designed to resist refutation by finkish dispositions or other counterexamples to the SCA. Choi argues that the intrinsic nature of dispositions doesn’t threaten either the SCA or the RCA.10
      Dispositions have been understood as persisting properties or states that make possible other properties or states, and they have been contrasted with occurrences, which have been understood as more episodic or transient events. Mumford (2011) says that the distinction between dispositions and occurrences derives mainly from the work of Gilbert Ryle (1949), and that as the appeal of Ryle’s form of behaviorism has faded, interest has shifted from the distinction between dispositions and occurrences to the distinction between dispositional and categorical properties.11
      Ryle (1949) distinguishes between dispositional terms (which may describe tendencies, propensities, potentialities, abilities, etc.) and occurrent terms (which may describe episodic or actually occurring events). He argues that it is mistaken to assume that any term that has a dispositional use must also have a corresponding episodic use.12 For example, the dispositional terms “know” and “believe” may not have episodic counterparts, and the dispositional statements “I know” and “I believe” may not correspond to episodic acts of knowing or believing. (I can’t properly say, conceptually or grammatically, “I am, at this moment, knowing or believing such and such.”) Ryle thus argues that mental states are dispositions to behavior, rather than unseen and unobservable occurrences.
      Ryle also distinguishes between generic and specific dispositions. He offers the sentence, “He is a cigarette smoker” as an example of a disposition ascription, and the sentence, “He is smoking a cigarette now,” as an example of an occurrence ascription. To say that someone is a cigarette smoker is not say that he is currently smoking, but to say that he has a disposition or tendency to smoke. Cigarette smoking is a specific disposition, in contrast to more generic dispositions such as the disposition to seek a method of reducing stress, relieving pressure, managing anxiety, or producing euphoria. Dispositional properties, according to Ryle, are nothing more than habits, tendencies, or other behavior regularities.
      McKitrick (2003) explains that Ryle’s distinction between specific and generic dispositions also corresponds to the distinction between single-track dispositions (those that are triggered by only one kind of circumstance, and that have only one kind of manifestation) and multi-track dispositions (those that are triggered by more than one kind of circumstance, and that have more than one kind of manifestation).13
      Against Ryle’s view, it may be argued that there may indeed be occurrent, as well as dispositional, mental states. I may know or believe something at this very moment, and this knowing or believing may be an occurrent state.
      It may also be argued that dispositions themselves are occurrences. There may be occurrent dispositions and dispositional occurrences. Dispositions may not always be separable from occurrences.
      E.J. Lowe’s account (2006, 2009, 2013) of the dispositional-occurrent distinction may best be understood in the context of his four-category ontology. In this formal ontology, there are four fundamental ontological categories (or kinds of being): substantial universals (kinds), substantial particulars (objects), non-substantial universals (attributes), and non-substantial particulars (modes). Kinds (such as tomatoes) are characterized by attributes (such as redness), and are instantiated by objects (such as particular tomatoes). Attributes (such as redness) characterize kinds (such as tomatoes), and are instantiated by modes (such as the redness of a particular tomato). Objects instantiate kinds, and are characterized by modes. Modes characterize objects, and instantiate attributes.
      Lowe regards instantiation and characterization as the two fundamental or primitive metaphysical relations. Another metaphysical relation, that of exemplification, he does not regard as fundamental or primitive, since it may result from instantiation or characterization, and may come in two different varieties: dispositional or occurrent. An object (such as a particular tomato) dispositionally exemplifies an attribute (such as redness) when it instantiates some kind (such as tomatoes) that is characterized by that attribute, and it occurrently exemplifies that attribute (redness) when it (the particular tomato) is characterized by some mode (such as its own redness) that instantiates that attribute.14
      In addition to making a distinction between dispositional and occurrent exemplification, Lowe makes a distinction between dispositional and occurrent predication. The sentence, “This piece of salt is water-soluble,” is an example of the former, and the sentence, “This piece of salt is dissolving in water,” is an example of the latter.15 Universal terms (such as “salt”), as well as individual terms (such as “this piece of salt), may serve as subjects of dispositional and occurrent predication.16
      Martijn Blaauw (2013) distinguishes between dispositional and occurrent belief. An example of the former would be John’s belief that physical exercise is good for his health, and an example of the latter would be John’s belief, as he looks at his watch and sees that it is 3 o’clock, that it is indeed 3 o’clock. Blaauw also distinguishes the disposition to believe from dispositional and occurrent belief. For example, John may have a disposition to believe something, without ever having dispositionally or occurrently believed that thing.
      David Rose and Jonathan Schaffer (2013) argue that knowledge entails dispositional belief, since “if s knows that p, then s dispositionally believes that p.” Knowledge is not merely occurrent belief (conscious endorsement of a thought), but dispositional belief (information available to mind for endorsement).17
      The distinction between dispositions and occurrences may be complicated by the fact that there may be many different kinds of dispositions (moral, aesthetic, behavioral, cognitive, emotional, behavioral, social, and cultural), and many different kinds of occurrences (historical, geographical, environmental, evolutionary, physical, biological, physiological, genetic, epidemiological, and statistical).
      Another question to be considered is whether dispositions can conflict with one another. Thus, Choi (2012) asks whether opposing dispositions can be co-instantiated. He defines conflicting dispositions as dispositions that have mutually consistent characteristic stimuli but inconsistent manifestations, and he concludes that opposing dispositions cannot be co-instantiated by one and the same object at the same time. He admits, however, that there are opposing viewpoints regarding this question of whether a single stimulus can trigger the manifestation of opposing dispositions.18
      There may also be disagreement about whether some dispositions are innate or acquired, intrinsic or extrinsic, “natural” or “unnatural.”
      There may also be positive and negative dispositions (“dispositions to” and “dispositions not to”).
      Dispositions may vary in their duration, strength, stability, and susceptibility to change.  They may also vary in the consistency with which they are manifested.
      Occurrences may be variously described as frequent, occasional, rare, unusual, expected, unexpected, regular, irregular, variable, invariable, timely, untimely, concomitant, or coincidental.
      Jennifer McKitrick (2009) describes “dispositional pluralism” as the view that there are many different kinds of dispositions,19 and “dispositional essentialism” as the view that all dispositions are essential properties of the objects that instantiate them.20 Dispositional pluralism is compatible with property dualism (the view that there are two fundamental kinds of properties: dispositional and non-dispositional) and with pandispositionalism (the view that all properties are dispositional), but not with anti-dispositionalism (the view that no properties are dispositional).21
      McKitrick (2003) says that there are extrinsic, as well as intrinsic, dispositions. Intrinsic dispositions are intrinsic properties of the things that have them, and do not depend on what is going on outside of those things. Extrinsic dispositions, on the other hand, are extrinsic properties of the things that have them, and depend on what is going on outside of those things.22
      The dispositional-categorical distinction may be even more difficult to define than the dispositional-occurrent distinction. Dispositional properties may be categorical, in the sense that they may categorically (rather than hypothetically) belong to things. Categorical properties may be dispositional, in the sense that they may take the form of dispositions. Properties may have both dispositional and categorical aspects.
      On the other hand, dispositional properties may be conditional in a way that categorical properties are not.
      According to the identity theory of dispositional and categorical properties, dispositional properties are the same as categorical properties. All dispositional properties are categorical, and all categorical properties are dispositional.
      Stephen Mumford (1998) describes property dualism as the theory that there are two fundamentally different kinds of properties: dispositional and categorical. Categorical monism is the theory that all properties are categorical, and dispositional monism (pandispositionalism) is the theory that all properties are dispositional.23
      Mumford also describes four types of property monism. Categorical reductionism is the theory that all dispositional properties can be reduced to categorical properties. Dispositional reductionism is the theory that all categorical properties can be reduced to dispositional properties. Categorical eliminativism is the theory that all properties are categorical, and that all reference to supposed dispositional properties should be eliminated. Dispositional eliminativism is the theory that all properties are dispositional (even those traditionally regarded as paradigmatically categorical), and that all reference to supposed categorical properties should be eliminated.24


1Tim Crane, “Introduction,” in Dispositions: A Debate, by D.M. Armstrong, C.B. Martin, and U.T. Place, edited by Tim Crane (London: Routledge, 1996), p. 1.
2Stephen Mumford, “Dispositions,” in Routledge Encyclopedia of Philosophy, 2011,
3Jennifer McKitrick, “A Case for Extrinsic Dispositions,” in Australasian Journal of Philosophy 81:2 (June 2002), p. 157.
4Tim Crane, “Introduction,” in Dispositions: A Debate, p. 5.
5C.B. Martin, “Dispositions and Conditionals,” in Philosophical Quarterly, 44, (1994), 1-8.
6Mumford, “Dispositions,” in Routledge Encyclopedia of Philosophy, 2011.
7Sungho Choi, “Dispositional Properties and Counterfactual Conditionals,” in Mind 117 (2008), 795-841.
8David Lewis, “Finkish Dispositions” (1997) in Papers in Metaphysics and Epistemology (Cambridge: Cambridge University Press, 1999), pp. 133-151.
9Sungho Choi, “The Conditional Analysis of Dispositions and the Intrinsic Dispositions Thesis,” in Philosophy and Phenomenological Research, Vol. LXXVIII, No. 3, May 2009, p 571.
10Ibid., p. 568.
11Mumford, “Dispositions,” in Routledge Encyclopedia of Philosophy, 2011.
12Gilbert Ryle, The Concept of Mind (Chicago: The University of Chicago Press, 1949), p. 119.
13Jennifer McKitrick, Dispositional Pluralism,” in Debating Dispositions: Issues in Metaphysics, Epistemology, and Philosophy of Mind, edited by Gregor Damschen, Robert Schnepf, and Karsten R. Stüber (Berlin & New York: Walter de Gruyter, 2009), p. 188.
14E.J. Lowe, More Kinds of Being: A Further Study of Individuation, Identity, and the Logic of Sortal Terms (Chichester: John Wiley & Sons, 2009), pp. 10-11.
15Ibid., pp. 142-143.
16Ibid., p. 144.
17David Rose and Jonathan Schaffer, “Knowledge Entails Dispositional Belief,” in Philosophical Studies (2013), Vol. 166, Issue 1, Supplement, p. 22.
18Sungho Choi, “Can Opposing Dispositions be Co-instantiated?”, in Erkenntnis, Vol. 8, Issue 1, Feb 2013, pp. 161-1182.
19McKitrick, Dispositional Pluralism,” in Debating Dispositions: Issues in Metaphysics, Epistemology, and Philosophy of Mind, p. 186.
19Ibid, p. 193.
20Ibid., p. 201.
21McKitrick, “A Case for Extrinsic Dispositions,” in Australasian Journal of Philosophy 81:2 (June 2002), p. 158.
22Stephen Mumford, Dispositions (Oxford: Oxford University Press, 1998), pp. 18-19.
23Ibid., pp. 172-175.


Martijn Blaauw, “Contrastive Belief,” in Contrastivism in Philosophy, edited by Martijn Blaauw (New York: Routledge, 2013), pp. 90-91.

Sungho Choi, “What is a Dispositional Masker?” Mind (2011), Vol.120, Issue 480, pp. 1159-1171.

Sungho Choi and Michael Fara, “Dispositions,” 2012, in Stanford Encyclopedia of Philosophy,

E.J. Lowe, The Four-Category Ontology: A Metaphysical Foundation for Natural Science (Oxford: Oxford University Press, 2006).

E.J. Lowe, Forms of Thought: A Study in Philosophical Logic (Cambridge: Cambridge University Press, 2013).