Abstract
All C-moralities remain rather indeterminate until the notion of an ‘alternative action’ is clarified. It is not a trivial task to find a concept of alternatives that fits into a consequentialist theory. Even if one has full empirical knowledge and a satisfactory analysis of performability, there is no obvious answer to the question of what the relevant alternative actions are, from the point of view of C, for a given agent in a given situation. As has been shown by Lars Bergström and others, the intuitively appealing answer ‘every action that the agent can perform in this situation’, turns out to be more problematic than one might expect. (Nevertheless, this is the answer I shall eventually accept.)
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Ergströmo). In several later articles, notably Bergström(3), Bergström(6) and Bergström(7), he defends and further elaborates his proposal.
Cf. Bergström(I), p. 39.
Bergström(1), pp. 21–22. See also Chapter 8, below.
Bergström(1), p. 30.
According to Bergström “a particular action a is performable if, and only if, the agent of a has (or had, or will have) it in his power to perform a”. (Bergström(l), p. 28.) He acknowledges that the ‘have it in one’s power’-idiom is very vague and ambiguous, but he does not attempt any further analysis.
In the sequel, when I say that an action a is ‘performable by P in S’ I shall mean that a is immediately performable by P in S, or, in other words, that a′s first moment roughly coincides with the point of time corresponding to S (see 5.3).
Bergström(1), pp. 36–37.
Bergström(6), p. 125, and Bergström(7), p. 91. Cf. Bergström(I), p. 51, and Bergström(3), p. 249.
Bergström(1), p. 41, Bergström(6), p. 124. (Bergström uses the terms T-consistent’, and T-inconsis-tent’, since T’ symbolizes his general teleological theory.)
Bergström(1), p. 51, Bergström(7), p. 90.
Bergström(7), p. 90. Cf. Bergström(l), p. 51, where Bergström gives an equivalent definition in terms of ‘versions’ and ‘quasi-versions’, instead of ‘variants’.
Bergström(6), pp. 143–144.
Bergström(7), p. 91. U violates MA, since an action may fail to satisfy U, despite being uniquely best in some alternative-set.
Bergström(7), p. 91.
Sobel(4), pp. 115–116.
Sobel(4), p. 116, note 1. Sobel attributes this observation to Bergström.
Sobel(4), p. 116, note 1.
Rabinowicz has constructed a more complicated case where U yields incompatible prescriptions. He also suggests two amendments of U. (Rabinowicz(I), pp. 23–24.) These amended principles are not entirely convincing, though. Even that which makes the weakest requirement on obligatoriness (Rabinowicz says that the principles may be equivalent), seems very seldom to make any action obligatory. This principle says, in effect, that an action ought to be done iff it is the only action in the situation that is optimal in some alternative-set satisfying (R). In most situations, there are probably several such actions.
In Carlson, Section 6.2, I discuss and criticize some other criteria, suggested by Bergström in Bergström(I), pp. 51–53.
Bergström(1), pp. 31–32.
This was pointed out to me by Rabinowicz.
Bergström(7), p. 87.
Bergström(1), pp. 34–35.
Bergström rejects DC in Bergström(6), p. 140, and in Bergström(7), p. 94, but repeats his argument for the incompatibility-requirement in Bergström(6), pp. 122–123..
It is satisfied by my principles KO*-KW*, presented in the next section.
The ‘entailment’ spoken of here need not be logical, but only ‘practical’. The agent cannot perform an alternative to ~a’, without performing (a variant of) a’.
As a further argument for the incompatibility-requirement, Bergström cites the alleged fact that ‘alternatives’ is ordinarily used in senses like ‘mutually exclusive possibilities’. (Bergström(l), p. 33.) But if using ‘alternative’ in a sense that does not imply mutual exclusiveness would deviate too far from common usage, we could just replace it by some other term. Being concerned with ‘alternatives’ in exactly the ordinary sense of this term is hardly an essential feature of consequentialism.
Bergström admits that he has not found a fully satisfactory solution to this problem. (Bergström(l), p. 54, Bergström(8), p. 141.)
Cf. Bergström(l), p. 50.
Rabinowicz pointed out to me that abandoning the requirement of incompatibility may lead to violations of NI. He has also made me aware of a certain problem concerning this principle. If a conditional analysis of ability, in the style of CA7, is correct, there are performed actions which would not have been perform-able, had they not been performed. Suppose that Brown tries to lift a 100 pound weight, and succeeds. Then his action of lifting a 100 pound weight (at that particular time) is performable. But the nearest possible world where he does not lift this weight may well be a world where he wants and tries to lift it, but fails. A typical conditional analysis would then say that this action was not performable. Assuming that Brown’s lifting the weight is an obligatory action (and that ‘ought’ implies ‘can’), we get a violation of NI. This action would not have been obligatory if it had not been performed, since it would not then have been performable. The violation of NI in this case is father special, however, since it concerns an action which would not only lack moral status if it were not performed, but would not even be an alternative for Brown in this situation. Confronted with this type of case, we might weaken NI in the following way: NF If an action is an alternative for P in S whether or not it is performed, then its moral status does nottlepend on whether or not it is performed. The principles SNI and (I), below, could be weakened in a similar way.
There are also some other actions that lack moral status. An example is ~a, in Case 5, below.
Prawitz(2), p, 81, Prawitz(3), p. 117.
Prawitz(2), p. 81, Prawitz(3), p. 117.
Prawitz(2), p. 80.
Prawitz(3), pp. 117–118.
Cf. Prawitz(2), pp. 78–79, Bergström(2), pp. 165–166, and Bergström(7), pp. 84–85.
Bergström(2), p.169
I am indebted here to Sobel, who launches a similar objectio against a principle which is rather like T1. (Sobel(3), p. 196) These problem could be avoided by demanding that alternatives must be situation-identical. But T1 would then prescribe both a3 and a4 in Case 6. This is an eqully serious flaw.
Prawitz also suggests a principle T2, which he regards as in a certain respect more plausible than T1 (Prawitz(3), p. 124.) In Carlson, p. 106 I argue that T2 is in fact no improvement on T1
Prawitz(3), p. 125.
Prawitz(3), p. 125.
Bergstrom(6), p. 133. Bergström suggests a certain amendment of T3, which I discuss in Carlson, pp. 108–109.
The same point is made by Bergström, in Bergström(2), p. 170, and in Bergström(6), pp. 135–136. Admittedly, KO*-KW* could also be regarded as ‘time-relativized’ (or ‘situation-relativized’), in one sense, since they allow that mutually incompatible actions are obligatory in different situations. But this kind of ‘time-relativity’ is not objectionable, I think.
ÅQVIST, p. 308.
Åqvist, p. 304. My notation is not identical with Åqvist’s.
Åqvist, p. 310. Prawitz mistakenly reports that Åqvist accepts the incompatibility-requirement. (Prawitz(3), p. 124.)
ÅQVIST, p. 310. In stating these principles, I have deviated considerably from Åqvist’s own concepts and notation.
Prawitz(3), p. 124, Bergström(6), p. 129.
Cf. Åqvist’s Definition 7. (Åqvist, p. 318.) Concerning courses of action, Åqvist uses the term ‘open’, instead of ‘performable’. His definition of ‘openness’, for a course of action at a time, is found in ÅQVIST, pp. 316–317.
ÅQVIST, p. 318.
Some further principles which Åqvist suggests (Åqvist, pp. 311–312) are discussed in CARLSON, pp. 111–112.
Björnsson.
Björnsson, p. 28.
Björnsson, pp. 28–29.
I here suppose that ‘outcome’ is interpreted as in WO or FO. Björnsson endorses a different conception of outcomes, based on J.L. Mackie’s analysis of causality, but I believe that violations of SNI will occur also according to that conception. (See Björnsson, p. 11, note 8.)
A slightly different example is needed to show that Case 8 is possible according to CA7. I give such an example in CARLSON, p. 117.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1995 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Carlson, E. (1995). Alternatives and Consequentialist Principles. In: Consequentialism Reconsidered. Theory and Decision Library, vol 20. Springer, Dordrecht. https://doi.org/10.1007/978-94-015-8553-8_6
Download citation
DOI: https://doi.org/10.1007/978-94-015-8553-8_6
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-4571-3
Online ISBN: 978-94-015-8553-8
eBook Packages: Springer Book Archive