The book Normativity by Judith Jarvis Thomson seeks to redefine the notion ‘good’ and many related concepts in philosophy. She discusses goodness, virtues, ought, reasons and correctness of mental states and acts. Here, I will focus on goodness. Thomson sets up a meta-ethical discussion about the nature of the word ‘good’. Her analysis is best described as a linguistic study from which she derives what ‘good’ means.
A first step in her analysis is the distinction between two kinds of adjectives (adj), predicative and attributive adjectives:
(i) Thomson defines predicative adjectives as
a. A is an adj E1.
b. A is an E2.
c. Entails: A is an adj E2.
For example ‘wooden’ is a predicative adjective: (a) A is a wooden house. (b) A is a barn. (c) Entails: A is a wooden barn.
(ii) Attributive adjectives are all those adjectives that are not predicative. For example, ‘heavy’ is an attributive adjective.
a. Fred is a heavy cat.
b. Fred is an animal.
c. Does not entail: Fred is a heavy animal. (Elephants, cows or hippos are heavy animals.)
Based on this distinction Thomson argues that ‘good’ is an attributive adjective. Her arguments starts from professions or roles that people play:
a. Alex is a good tennis player.
b. Alex is a chess player.
c. Does not entail: Alex is a good chess player.
Easily her argument can be applied for goodness of artifacts:
a. A is a good printer.
b. A is a scanner.
c. Does not entail: A is a good scanner.
From the attributive nature of ‘good’, Thomson derives that goodness as such does not exist. She emphasizes that just stating that something is good is ambiguous. Saying "This car is good." would make many people believe that the car would be reliable, consumes little fuel, or drives fast. However, some other people might think of safety when calling a car good. And even a very poor car, which is poor in everything, would at least be good in being poor. Therefore, Thomson concludes that simply stating "A is good" does not provide any information about A, even though we often use it in such ways.
In order to give meaning to the notion of goodness, Thomson derives four goodness properties for a kind K:
(I) being a good Kgf
(II) being good-modified
(III) being good-modified for a K
(IV) being a good Kgf for a K
In order to derive (I), Thomson argues that all good things are only good, respectively the kind K they belong to. Tennis players, tigers, Beefsteak tomatoes, physicians, etc. are called good for different reasons. Indeed "each of them is such that what being a K is itself sets the standards that a K has to meet if it is to be good qua K." (p.21) If a K fulfills these standards, Thomson calls it a good-fixing kind (Kgf).
'Being good-modified' refers to goodness in particular respects, e.g. 'being good for', 'being good as', 'being good in'. For instance, 'being good in crossword puzzles', 'being good for England', 'being good with children'.
The third goodness property is a restriction of the second on a particular kind. For example, 'Tom is good in chess for a child' is good modified and restricted to children. Overall, Tom might lose most of his parties (thus overall he is not good in chess), while being the best in his age (good in chess for a child).
Derived from this goodness property is property (IV). Similar to restricting 'being good-modified' on a particular kind, one can also restrict goodness-fixing kinds on particular kinds. For example, 'Tom is a good chess player for a child' means exactly this. Here Tom is a good-fixing kind restricted to the kind children.
The biggest problem of Thomson is the potential ambiguity of 'good'. It is imaginable that one synonym of 'good' is an attributive adjective, and another synonym is a predicative attributive; 'good as such' would be such a predicative adjective. First she attacks Moore's and Ross's usage of 'goodness'. Her argument against Moore is not convincing. Thomson claims that there are only three ways of understanding goodness: the Goodness Thesis -- goodness is the property that all good things have in common; Alternate Rationale -- when calling something good, we ascribe to it the property goodness; and, Thomson's view of 'good' as an attributive adjective. Then she argues why the first two positions are difficult to defend and concludes that her position is plausible. This conclusion however is clearly not sufficiently strong for excluding other synonyms of 'good'. In fact, there is no argument given why the three positions are exclusive, nor why any other position than Thomson's would be self-contradictory.
Second, she gives arguments against Ross's interpretation of goodness. Ross claims that the philosophical most important synonym for goodness is categorical goodness, e.g. "Pleasure is good" or "Knowledge is good." Again Thomson asks what such concept of goodness could mean, provides a list of possible meanings, and rules them out. Again, she does not provide an argument that the list would be complete or that any predicative use of 'good' leads to a contradiction.
Finally, it remains the question whether Thomson's notion of goodness can cover goodness as proposed by moral theories. For virtue ethics, her notion clearly can. She provides a chapter on virtues. However, for consequentialism or deontological ethics this is not so clear. Thomson attempts to deny that consequentialism is a sound moral theory at all. Unfortunately her major attack on consequentialism in chapter IV remains general and repeats well known arguments.
In summary, Thomson's book Normativity provides a well based account of instrumental goodness. Since the account also works well as basis for virtue ethics, it raises questions about the similarities of virtues in persons and artifacts. Unfortunately, Thomson's attack on the established notion of goodness is underdeveloped. After reading her book, one is not tempted to believe that goodness in her sense is the only possible synonym in which goodness can be used.
Judith Jarvis Thomson (2008). Normativity, Open Court Publishing: Chicago/La Salle.
No comments:
Post a Comment