Contemporary Approaches to Causation

  • Menno Hulswit
Part of the Philosophical Studies Series book series (PSSP, volume 90)


This chapter consists of two parts. In the first section, I will give a general outline of the current approaches to causation. In the second, I will discuss a number of pertinent problems the current approaches are afflicted with.


Short Circuit Causal Relation Causal Explanation Causal Statement INUS Condition 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    There is an enormous amount of recent literature on the problem of causation. I have found Sosa and Tooley (1993), Taylor (1967), and Kim (1995) especially helpful in coming to an overview of the contemporary discussions.Google Scholar
  2. 2.
    Borrowed from Taylor 1966.Google Scholar
  3. 3.
    This is a slightly changed version of the definition given by Sosa en Tooley 1993, 6.Google Scholar
  4. 6.
    Sosa and Tooley (1993, 6-7) would not accept our interpretation; they defend the idea that causal sufficiency entails logical necessity.Google Scholar
  5. 7.
    Taylor (1966, 35-7) gives several interesting examples of ‘undeterminative sufficiency.’Google Scholar
  6. 8.
    The counterfactuals in turn are explained on the nomic-inferential model. It is at this point that laws and regularities enter into singular causal judgments. According to Mackie, his analysis &quote;has been given entirely within the limits of what can still be called a regularity theory of causation&quote; (Mackie [1965] 1993, 52).Google Scholar
  7. 9.
    A problem, put forward by Kim (1971; also 1973), that is more specific to the INUS condition approach is that its implicit ontology of events must be carefully worked out. According to Kim, &quote;any discussion of causation must presuppose an ontological framework of entities among which causal relations are to hold, and also an accompanying logical and semantical framework in which these entities can be talked about&quote; (Kim [1971] 1993, 60). Apparently, Mackie often confuses events with their descriptions (Kim [1971] 1993,68).Google Scholar
  8. 10.
    Lewis mentions some problems that were not listed here. According to him, the main problems for the regularity analysis are (a) distinguishing genuine causes from effects, (b) epiphenomenona, and (c) pre-empted potential causes: &quote;regularity analyses tend to confuse causation itself with various other causal relations. If c belongs to a minimal set of conditions sufficient for e, given the laws, then c may well be a genuine cause of e. But c might rather be an effect of e: one which could not, given the laws and some of the actual circumstances, have occurred otherwise than by being caused by e. Or c might be an epiphenomenon of the causal history of E: a more or less inefficacious effect of some genuine cause of e. Or c might be a pre-empted potential cause of e: something that did not cause e, but that would have done so in the absence of whatever really did cause e&quote; (Lewis [1973] 1993, 193-94).Google Scholar
  9. 13.
    According to Lewis, causal dependence among events implies causation, but the converse does not hold: &quote;If c and e are two actual events such that e would not have occurred without c, then c is the cause of e. But I reject the converse. Causation must always be transitive; causal dependence may not be; so there can be causation without causal dependence. Let c, d and e be three actual events such that d would not have occurred without c and e would not have occurred without d. Then c is a cause of e even if e would still have occurred (otherwise caused) without c&quote; (Lewis [1973] 1993, 200).Google Scholar
  10. 14.
    In Lewis’s own words: &quote;one event is a cause of another iff there exists a causal chain leading from the first to the second&quote; (Lewis [1973] 1993, 200).Google Scholar
  11. 15.
    Horwich [1987] 1993, 208-11.Google Scholar
  12. 16.
    There has been a historical development from I to II to III. Strictly speaking, sense I is the only proper sense. Whereas I and II originated in ancient Greece, III was a Renaissance invention. According to I, causation entails that a free and responsible agent is influenced by another agent who provided him with a motive for doing something. If we read in the newspaper &quote;Mr. president’s speech causes havoc in parliament,&quote; this does not mean that the president compelled the members of the parliament to quarrel; it rather means that his speech afforded them a motive for doing so. A cause in sense II is something that can be used to manipulate things in nature. Causes I an II are contingent in a double sense: they are contingent in their existence because their existence depends on human volition. They are contingent in their operation because in their operation they depend for the production of their effects on further conditions. Cause III is a condition or set of conditions that are invariably accompanied by some change. Causes III are necessary both in their existence and in their production. Their existence does not depend on human beings, and the production of their effects does not depend on any further conditions. There can be no relativity of causes, and no diversity of effects depending on other conditions. There is a one-one relation between cause and effect (Collingwood [1938] 1991, 145-54).Google Scholar
  13. 17.
    His approach to the problem of directionality (or causal priority) was: &quote;I now propose the following way of distinguishing between cause and effect by means of the notion of action: p is a cause relative to q, and q an effect relative to p, if and only if by doing p we could bring about q or by suppressing p we could remove q or prevent it from happening&quote; (Von Wright 1971, 70).Google Scholar
  14. 18.
    For a much more detailed criticism of Gaskin’s and Von Wright’s accounts, see Tooley 1987,239-242.Google Scholar
  15. 19.
    D.H. Mellor (1986, 1995) takes the relata of the causal relation to be facts rather than events.Google Scholar
  16. 20.
    There is some disagreement about the relationship between causation and probabilistic dependence: whereas, for example, Lewis (1986) takes probabilistic dependence to be sufficient, but not necessary, for causation, Mellor (1986) takes probabilistic dependence to be both necessary and sufficient for causation.Google Scholar
  17. 21.
    Mackie 1974, 134-142 contains an extensive critique of Ducasse’s view.Google Scholar
  18. 22.
    More precisely, see Sosa and Tooley 1993, 1-5, 11-14, and Tooley 1987, 173-8, 183-190, 202-204, and 244-50.Google Scholar
  19. 23.
    Until recently, it was generally thought that Hume held the analytical reductionist position. A number of recent philosophers, however, have argued that Hume was a realist rather than a reductionist. For explanation and further references, see Galen Strawson 1989.Google Scholar
  20. 24.
    A. N. Whitehead (1920, 1929) (and the philosophers of process in general) seems to be a remarkable exception to the rule mentioned by Sosa and Tooley; though he insists that we do have an immediate experience of causation, he also insists that it is one of the main tasks of any metaphysics to give an adequate analysis of this experience. Whereas Sosa’s and Tooley’s interest is primarily epistemological-an analysis of the concept of causation, Whiteheads interest is primarily metaphysical-an analysis of our experience of causation.Google Scholar
  21. 25.
    Though Tooley does not explain (explicitly) what he means by a theoretical relation, he tells us that &quote;to be a realist with regard to theoretical entities is to hold that facts concerning the relevant unobservable entities (or events, etc.) are not logically supervenient upon observable facts&quote; (Tooley 1987, 177). Whereas Tooley holds that causal relations are not logically supervenient upon non-causal facts (1987, 177), he insists that &quote;any knowledge that we have of causal connections must ultimately rest upon knowledge of non-causal states of affairs&quote; (Tooley 1987, 247). Causal terms are analyzable &quote;by means of any method that is applicable to theoretical terms in general.&quote; [...] &quote;Moreover, given that causal terms are theoretical, there is no need to postulate any special faculty that provides one with direct awareness of causal relations. Causal claims will be epistemologically justified in the same way that theoretical claims in general are justified&quote; (Tooley 1987, 249-50).Google Scholar
  22. 26.
    The term ‘event’ is ambiguous: it may refer to a particular occurrence, happening only once, with a particular duration and location. These are called ‘token events.’ There are, however, also ‘type events,’ such as, for example, the Olympic Games, which happen every four years. Most contemporary philosophers are predominantly concerned with events as particulars. Chisholm (1970) is an exception; he takes events to be universals, capable of recurrence.Google Scholar
  23. 27.
    A notable exception is Emmet 1984, 1992.Google Scholar
  24. 28.
    Davidson’s idea that causal explanations need not refer to laws in the strong sense of the word, is clearly expressed by Dorothy Emmet: &quote;When we speak of a singular event as caused or as having effects, this is said to be explanatory when it can be connected with some generalisation-this need not be as elevated as a Law; it can be a commonsensical generalisation as that ships which get great gashes in their holds so that water pours in are likely to sink (the law would be that of the specific gravity of water). So the singular event has a property or properties by which it can be assigned to a class of events, and it is followed by another event whose properties also assign it to a class, and a causal generalisation can be made by stating a relation between properties of events of these two classes. The causal statement about a particular event is being made about it as an instance of a kind, and under the description which assigns it to the kind named by these properties&quote; (Emmet 1984, 20-21).Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2002

Authors and Affiliations

  • Menno Hulswit
    • 1
  1. 1.Heyendaal InstituteUniversity of NijmegenThe Netherlands

Personalised recommendations