Abstract
Geach (Analysis 17: 33–42, 1956) and Thomson (J Philos 94:273–298, 1997, Normativity, 2008) have argued that nothing is just plain good, because ‘good’ is, logically, an attributive adjective. The upshot, according to Geach and Thomson, is that consequentialism is unacceptable, since its very formulation requires a predicative (non-attributive) use of ‘good’. Reactions to the argument have, for the most part, been uniform. Authors have converged on two challenging objections (Ross, The right and the good, 1930; Pidgen, Philos Q 40:129–154, 1990; Arneson, Analysis, 70:731–744, 2010; Smith, Analysis 70:715–731, 2010; Sturgeon, Analysis 70:744–753, 2010; Kraut, Against absolute goodness, 2011). First, although the logical tests that Geach and Thomson invoke clearly illustrate that ‘good’, as commonly used, is an attributive, they don’t show that ‘good’ lacks an intelligible predicative interpretation. Second, even if the English word ‘good’ fails to express the property of goodness, we can just stipulate that ‘good*’ expresses goodness and thus formulate consequentialism accordingly. The second objection is one way of voicing skepticism about the method of drawing substantive philosophical conclusions from considerations about ordinary language. In this essay, we present an argument, inspired by Geach and Thomson, which isn’t susceptible to the same objections but which supports the same conclusion. The significance of our argument for ethics is obvious; it challenges the intelligibility of standard consequentialism, and even certain forms of non-consequentialism. One might be inclined to think that a more sophisticated consequentialism, which relies on ‘good {possible world/state of affairs/outcome}’ instead of just ‘good’, evades the criticism. But we explain why the criticism can’t be so easily evaded.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Geach is often cited as the originator of the observation that ‘good’ is attributive but it was anticipated by Ross (1930), who made rather less of it than Geach did. According to Kamp (1975, p. 127), the observation is so old that it can’t be precisely traced to its origin. One respondent informed us that essentially the same point can be found in Aristotle. More recently, the observation reappears in Ziff (1960), von Wright (1963), and Foot (1985).
This explanation is also given in the linguistics literature. See Kennedy (2012, p. 334). We should note that our use of ‘predicative’ and ‘attributive’ diverges slightly from the way these terms are used in linguistics. Linguists typically use them to distinguish between genuine predicates and expressions that happen to modify a noun phrase. We understand the claim that ‘good’ is sometimes predicative to amount to the thesis that ‘good’ sometimes expresses a property, goodness, which can then be predicated of various kinds of thing.
Authors who either defend this form of consequentialism, or take it seriously enough to discuss it, include Moore (1903), Dewey and Tufts (1908, pp. 224–225), Broad (1930, pp. 206–207), Ross (1930), Frankena (1963, p. 13), Rawls (1971, p. 24), Smart (1973, p. 9), Scheffler (1988, p. 1), Kagan (1998, p. 26), Brown (2011, p. 751), Parfit (2011, pp. 235–236), Temkin (2012, pp. 24–25), and Willenken (2012, p. 545).
Our formulation of standard consequentialism in terms of goodness presupposes realism about properties. The presupposition is mainly for expository purposes; it makes our job a bit easier. Of course, we acknowledge that a consequentialist may well embrace nominalism. Anyone who accepts both consequentialism and nominalism will think (roughly) that what one ought to do is produce more true instances of ‘a is good’, where the truth of the instances isn’t susceptible to further analysis in terms of the subject’s standing in some relation (instantiation) to an abstract entity such as goodness (Cf. Quine 1948 and Devitt 1980). Just as the nominalist about color properties believes that there are red things (particular houses, roses, sunsets) without there being some further entity, redness, to which they stand in the instantiation relation, the value nominalist believes that there are good things (perhaps houses, roses, sunsets) without there being some further entity, goodness, to which they stand in the instantiation relation. As we explain in Sect. 3, the use of ‘good’ in the phrase ‘good thing’ is a predicative use of the word.
Francis Kamm, for instance, appears to endorse a non-consequentialist response to value: “It is not our duty to bring about […] agent-neutral value, but only to respect the constraints that express its presence” (Kamm 1992, p. 386). Korsgaard (1983) and Langton (2007) have discussed creating and responding to value in a Kantian theory. See also Schapiro (2001) for discussion of different responses to value in utilitarianism, rational intuitionism, and Kantian constructivism. Whether Kant himself was a non-consequentialist of the sort we’re describing is, we believe, a controversial question.
‘Famous’ is discussed in Thomson (2008, p. 14) and credited to Matthew Hanser. In order to distinguish ‘good’ from ‘famous’ in such a way as to rescue her argument, Thomson asks a rhetorical question: “What assures us that ‘famous’ does have this second use is that we know what the property of being (simply) famous is—it is the property of being (simply) well known. What is the property that Ross claims is ascribed to knowledge, or pleasure, by a philosopher who says ‘That is good’ of it?” The implication is that our inability to say, in other terms, what it is for something to be (simply) good should undermine our confidence in the intelligibility of predicative ‘good’. But this move is unlikely to persuade Thomson’s interlocutors. There are several reasons why.
First, it may well be that ‘good’, when used predicatively, is unanalyzable. Moore (1903), arguably the most important consequentialist in the twentieth century and the author to whom Thomson addresses her arguments, explicitly defends the unanalyzability of predicative ‘good’. Furthermore, some contemporary writers are sympathetic to Moore’s view (see Kraut (2011, p. 182)). Second, Thomson’s argument overgeneralizes. If an inability to analyze predicative ‘good’ tells in favor of its unintelligibility, why then should not our inability to analyze ‘truth’, ‘existence’, etc. tell in favor of their unintelligibility? Finally, if Thomson’s response to the challenge involving ‘famous’ were successful, it would have rendered the initial argument in terms of Geach’s observation unnecessary.
We intend for (P1)–(P3) and (P5) to be interpreted as material conditionals.
We’re alluding, of course, to the Knight Industries Two Thousand (‘KITT’ for short)—the car depicted in the 1980s TV series, Knight Rider, starring David Hasselhoff.
Tomás de Torquemada was Grand Inquisitor of Spain in the fifteenth century, and oversaw the torture, forced conversion, and ultimate expulsion of many Spanish Muslims and Jews.
Mightn’t the predicative reading of ‘good’ just be the reading made salient in C? No, not if one takes the meaning of predicative ‘good’ to apply univocally across a wide range of kinds. For the property of being a morally good person isn’t the property that consequentialists want to ascribe to events, states of affairs, possibilities, or sensations, though they do often speak as if bare occurrences of ‘good’ expressed one and the same property—goodness—that particular persons, events, states, possibilities, and sensations can have. The reading of ‘good’ in C on which (A4) is valid isn’t the sort of reading that can modify so wide a range of kinds.
David Lewis rejects Kripke’s judgment about the rigidity of ‘pain’ by committing just this sort of error. He writes,
If the concept of pain is the concept of a state that occupies a certain causal role, then whatever state does occupy that role is pain. If the state of having neurons hooked up in a certain way and firing in a certain pattern is the state properly apt for causing and being caused, as we materialists think, then that neural state is pain. But the concept of pain is not the concept of that neural state. (“The concept of…” is an intensional functor.) The concept of pain, unlike the concept of that neural state which in fact is pain, would have applied to some different state if the relevant causal relations had been different. Pain might have not been pain. The occupant of the role might have not occupied it (1980, p. 125, emphasis added).
Now it’s certainly true that the occupant of the pain-role might have not occupied that role, but that’s a true interpretation of the sentence ‘pain might have not been pain’ only if the italicized clause in the passage is indeed true—only if, that is, the concept of pain is in fact the concept of a state that occupies a certain causal role. But that’s an extremely controversial and not at all obvious claim—one that we should abstract away from in passing a semantic judgment about ‘pain might have not been pain’, since the judgment is supposed to be one source of evidence about what sort of concept the concept of pain is. If analytic functionalism is true, then ‘pain might have not been pain’ has a coherent reading—granted—but that’s partly what’s at issue when Kripke asks us to assess the sentence, ‘pain might have not been pain’, for coherence.
As we indicated at the beginning, this is essentially G&T’s approach to analyzing such constructions. But, to emphasize, while we’re somewhat sympathetic to it, our central argument, unlike G&T’s, doesn’t rely on contextualism about ‘good’. In fact, the contextualist strategy may seem much more promising now that we have in hand an argument that the relevant expressions can’t express unqualified goodness.
According to one recent and highly attractive formal account, all gradable adjectives, like ‘expensive’ and ‘good’, are doubly context sensitive: context supplies a dimension along which the item is said to be expensive or good, and a threshold along a certain scale determined by the relevant dimension such that items on the right side of the threshold are expensive or good (Kennedy 2007).
For a recent example, see Temkin (2012, pp. 24–25). Temkin’s book is in fact called Rethinking the Good.
It’s important to keep in mind at this stage how our use of ‘attributive’ diverges from its standard use in linguistics. See footnote 2.
Geach (1956, p. 34) made this point as regards the phrase ‘good thing’.
Recall footnote 4, where we acknowledged the possibility of combining consequentialism with nominalism. The value nominalist, as we called her, maintains that there are good things but there isn’t any such thing as goodness. In light of our observations above, value nominalism can’t evade our objection: a commitment to the existence of good things involves a commitment to a meaningful use of predicative ‘good’. Our arguments in defense of (P4) demonstrate that there isn’t any such use of ‘good’.
Of course, it always sounds somewhat stilted to speak of something being good as a K, and so on. But ‘good {as a/for a/in its capacity as a/insofar as its a/qua} possible world’ is bizarre in a way that makes it difficult to understand.
A classic source for this view is Taurek (1977). In stating it we’ve assumed, for simplicity, that W1 and W2 are populated by the same inhabitants.
Thomson doesn’t discuss the property of being good for people in Normativity. However, her criticism of the view that one must promote ‘happier worlds’ can be adapted, which is roughly what we do here (2008, pp.61–67). .
References
Arneson, R. (2010). Good period. Analysis, 70, 731–744.
Broad, C. D. (1930). Five types of ethical theory. London: Routledge.
Brown, C. (2011). Consequentialize This. Ethics, 121, 749–771.
Devitt, M. (1980). “Ostrich nominalism” or “mirage realism”? Pacific Philosophical Quarterly, 61, 433–439.
Dewey, J., & Tufts, J. H. (1908). Ethics. New York: Henry Holt.
Foot, P. (1985). Utilitarianism and the virtues. Mind, 94, 196–209.
Frankena, W. K. (1963). Ethics. Englewood Cliffs: Prentice-Hall.
Geach, P. (1956). Good and evil. Analysis, 17, 33–42.
Hinzen, W. (2006). Mind Design and Minimal Syntax. Oxford: Oxford University Press.
Kagan, S. (1998). Normative ethics. Boulder: Westview Press.
Kamm, F. (1992). Non-consequentialism, the Person as an End-in-Itself, and the Significance of Status. Philosophy & Public Affairs, 21(4), 354–389.
Kennedy, C. (2007). Vagueness and grammar: The semantics of relative and absolute gradable adjectives. Linguistics and Philosophy, 30, 1–45.
Kennedy, C. (2012). Adjectives. In G. Russell, & D. G. Fara, (Eds.), Routledge Companion to Philosophy of Language. New York: Routledge.
Korsgaard, C. (1983). Two Distinctions in Goodness. Philosophical Review, 92, 169–195.
Kraut, R. (2011). Against absolute goodness. Oxford: Oxford University Press.
Kripke, S. (1980). Naming and necessity. Cambridge: Harvard University Press.
Langton, R. (2007). Objective and Unconditioned Value. Philosophical Review, 116, 157–185.
Lewis, D. 1980. Mad pain and Martian pain. Readings in the Philosophy of Psychology 1, 216–222. Reprinted in David Lewis, Philosophical Papers I, pp. 122–129. Page reference to this volume.
Moore, G. E. (1903). Principia ethica. Cambridge: Cambridge University Press.
Nunberg, G. (1978). The Pragmatics of Reference. Bloomington: Indiana University Linguistics Club.
Parfit, D. (2011). On what matters (Vol. 1). Oxford: Oxford University Press.
Pidgen, C. (1990). Geach on “good”. Philosophical Quarterly, 40, 129–154.
Quine, W. V. (1948). On what there is. The Review of Metaphysics, 2, 21–38.
Rawls, J. (1971). A theory of justice. Cambridge: Harvard University Press.
Rosen, G. (2012). Review of Judith Jarvis Thomson’s normativity. Journal of Philosophy, 109(11), 676–681.
Ross, W. D. (1930). The right and the good. Oxford: Oxford University Press.
Schapiro, T. (2001). Three conceptions of action in moral theory. Noûs, 35(1), 93–117.
Scheffler, S. (1982). The rejection of consequentialism. Oxford: Clarendon Press.
Scheffler, S. (Ed.). (1988). Consequentialism and Its critics. Oxford: Oxford University Press.
Smart, J. J. C. 1973. An outline of a system of utilitarian ethics. In Utilitarianism: For & against. Cambridge: Cambridge University Press.
Smith, M. (2010). On normativity. Analysis, 70, 715–731.
Sturgeon, N. (2010). Normativity. Analysis, 70, 744–753.
Taurek, J. (1977). Should the numbers count? Philosophy & Public Affairs, 6, 293–316.
Temkin, L. (2012). Rethinking the good. Oxford: Oxford University Press.
Thomson, J. J. (1997). The right and the good. Journal of Philosophy, 94, 273–298.
Thomson, J. J. (2008). Normativity. Chicago: Open Court.
von Wright, G. H. (1963). The varieties of goodness. London: Routledge & Kegan Paul.
Willenken, T. (2012). Deontic cycling and the structure of commonsense morality. Ethics, 122, 545–561.
Ziff, P. (1960). Semantic analysis. Ithaca: Cornell University Press.
Acknowledgments
We’re fortunate to have many supportive friends, colleagues, and teachers. This project was much improved by their questions and comments. We would like to thank Dominic Baily, David Barnett, James Bondarchuk, Ross Cameron, Fabrizio Cariani, Jennifer Carr, Rebecca Chan, Daniel Elstein, Ephraim Glick, Aidan Gray, Sam Fleischacker, Chris Heathwood, Ulrike Heuer, Dave Hilbert, Paul Hovda, Kathrin Koslicki, Tony Laden, Heather Logue, Connie Meinwald, Bernhard Nickel, Graham Oddie, Ben Rohrs, Noël Saenz, Paolo Santorio, Sally Sedgwick, and Robbie Williams. One person deserves special recognition. We were first introduced to the question, is anything just plain good, in the Fall of 2005 at MIT, when Judith Jarvis Thomson taught a graduate seminar about normativity. Our thoughts have been heavily influenced by her work. We’re extremely grateful to have had her as a teacher, and for her written comments on an earlier draft of this paper (though all remaining mistakes are our own). Finally, one author wishes to acknowledge the College of Liberal Arts & Sciences at UIC, which supported the final stages of this project by means of an LAS Award for Faculty Research in the Humanities.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Almotahari, M., Hosein, A. Is anything just plain good?. Philos Stud 172, 1485–1508 (2015). https://doi.org/10.1007/s11098-014-0361-5
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-014-0361-5