An Adamsian Theory of Intrinsic Value
- First Online:
- Cite this article as:
- Hill, S. Ethic Theory Moral Prac (2011) 14: 273. doi:10.1007/s10677-010-9244-9
- 92 Views
In this paper I develop a theological account of intrinsic value drawn from some passages in Robert Merrihew Adams’ book Finite and Infinite Goods. First I explain why Adams’ work on this topic is interesting, situate his theory within the broader literature on intrinsic value, and draw attention to some of its revisionist features. Next I state the theory, raise some problems for it, and refine it in light of those problems. Then I illustrate how the refined theory works by showing that it has the resources to deal with some seemingly formidable objections.
KeywordsAdams, R.M.Instrinsic valueGod
1 Establishing Interest
There are several reasons to be interested in Adams’ discussion of intrinsic value. First, it has proven difficult to give a plausible account of intrinsic value.1 Given these difficulties, and given Adams’ much deserved stature within the discipline, his work on this topic merits an examination.
Second, Adams’ theory of intrinsic value occupies the center of his ethical and metaethical thought. The wide number of otherwise independent projects discussed in his books2Finite and Infinite Goods: A Framework for Ethics and A Theory of Virtue: Excellence in Being for the Good are linked by their shared dependence on Adams’ theory. Given its importance to his other work, Adams’ theory of intrinsic value deserves to be examined.
Third, various sections of Finite and Infinite Goods have received praise from a wide and diverse range of prominent authors.3 Since so many authors, of philosophical persuasions very different from Adams’ own, have found so much to praise in Adams’ book, and since Adams’ theory of intrinsic value is so central to the overall project of that book, his theory merits a critical examination.
There are some respects in which Adams’ theory of intrinsic value is revisionist. Traditional ethicists attribute a certain structure to intrinsic value. Adams’ theory does not have that structure. In order to avoid confusion, it is important to identify these revisionist features of Adams’ theory before discussing its precise details. As I see it, there are three important respects in which Adams’ theory departs from tradition.
Intrinsic value is not intrinsic
Tradition holds that intrinsic value is an intrinsic property. Adams holds that it is a relational property.
Intrinsic badness is derivative
Tradition holds that intrinsic goodness and intrinsic badness are both fundamental. Adams holds that only intrinsic goodness is fundamental. All sorts of badness, including intrinsic badness, are derivative and are to be analyzed in terms of fundamental intrinsic goodness.
Extrinsic value alters intrinsic value
Tradition holds that the extrinsic value of an object cannot alter that object’s intrinsic value. The version of Adams’ theory that I offer here allows that the extrinsic value of an object can alter that object’s intrinsic value.
The phrase “intrinsic value” is something of a philosophical term of art. It is not, I think, an expression in much use in ordinary conversation or writing, not even among the fairly educated. Yet, at the same time, it also seems to me plausible to suggest that when philosophers introduce the term “intrinsic value” they are attempting to provide a label for a concept that does occur in ordinary thought, even if it only occurs implicitly and without a common label.... I say all this not so much to make a fetish out of ordinary beliefs about the concept of intrinsic value, but rather to warn us against the opposite danger, that is, that we will make a fetish out of philosophical beliefs about the concept of intrinsic value. In particular, it seems to me that the very label we have provided ourselves—“intrinsic value”—reflects a philosophical theory about the nature of the value in question. And it seems to me that this theory may well be false. Nonetheless, there is a strong temptation to think that the philosophical theory in question must be true. Realizing that the phrase “intrinsic value” is indeed a philosophical term of art, and given that the theory is effectively “built in” to the term itself, there is a strong temptation to think that the theory must be true, by definition. It is this temptation that I am especially keen to resist.
Throughout the paper I will be explicit about where Adams’ theory and my refinements of it depart from tradition. Some of these departures are inherited. In such cases my defense will simply involve an appeal to some passage in which the relevant departure is explicitly endorsed by Adams. A few of these departures are not endorsed by Adams. I either advocate an idea that he never explicitly discusses or I take one of his ideas further than he does. In such cases I will justify these violations of tradition by appealing to recent work on intrinsic value and to theoretical advantages purchased by such departures.
3 Adams’ Theory of Intrinsic Value
GOOD: An object, o, is intrinsically good to degree, n, if and only if o is similar to God to degree n.
Adams does not accept this simple formulation of the theory. There are examples that he thinks it gets wrong. Consider Hitler. Hitler was powerful. God is powerful. Hitler’s power, therefore, contributed to the degree to which he was similar to God. So if GOOD is true, then Hitler’s power contributed to the degree to which he was intrinsically good. But Hitler’s power did not contribute to the degree to which he was intrinsically good. So, GOOD is false.
Probably the best reply we could give to these counterexamples [such as the Hitler example] if we want to defend the theory that excellence consists simply in resembling God is that not every sharing of a property constitutes a resemblance. Judgments of resemblance are more holistic than that.... The way or context in which a property is shared affects whether the sharing constitutes a resemblance or makes things more similar than they would otherwise have been.... There is a more pressing difficulty, however, for the thesis that Godlikeness is sufficient for excellence. More decisive counterexamples to it are possible involving holistic resemblance rather than mere sharing of properties. Consider the phenomenon of parody or caricature. Parodies and caricatures do resemble, but do not in general share the excellences of their original or object.... It is natural enough to say that Hitler’s power is “a caricature of the divine power”—more natural, I suspect, than to deny flatly that his power resembles God’s in any way. This suggests a modification of the analysis of excellence in terms of resembling or imaging God. We can suppose that the difference between resemblances to God that do and do not constitute virtues or excellences is analogous to the difference between good portraits (by which I mean faithful portraits) and caricatures. The excellence of other things besides God will consist, then, in the faithfulness of their imaging of God. This cannot be more than an analogy, of course, due to the radical imperfection of any resemblance of creatures to God. I will not offer here a full account of what the faithfulness of a portrait amounts to, and I am not sure that I could give one. It would surely include the observation that caricatures are distorted in a way that faithful portraits are not. The caricature exaggerates one or more features of the original, whereas the faithful portrait represents features in a balanced way and in relation to those other features to which they are most importantly related in the original. How much, and what, must be included in a faithful image depends on what is most important about the way in which the original has the features shared or represented.
Here Adams makes two revisions to GOOD. First, GOOD employs an account of similarity according to which any property that an object shares with God contributes to the degree to which that object is similar to God. But, in the present passage, Adams claims that similarity is “holistic.” Merely sharing a single property, he says, is insufficient for similarity.
GOOD*: An object, o, is intrinsically good to degree, n, if and only if o faithfully resembles God to degree n.
Adams introduces the requirement about faithfulness in order to deal with the Hitler Objection. Perhaps this is the response he has in mind: Hitler was powerful. God is powerful. But merely sharing properties with God does not contribute to similarity to Him. Perhaps, in the relevant holistic sense, Hitler was similar to God. But Hitler was a caricature of God. He shared some properties with God, such as powerfulness; but the way in which Hitler had those properties was distorted, exaggerated, and unbalanced. So, GOOD* delivers the judgment that Adams wants—in spite of his great power, Hitler was not intrinsically good to a very large degree.
Adams’ reply to the Hitler Objection seems plausible when the judgment GOOD* delivers about Hitler is contrasted with the judgments it delivers about other subjects.
Jones: Jones is a very loving person. He is dedicated to the poor and spends a lot of time and effort trying to help them. Unfortunately, though, Jones is almost completely powerless. In spite of this, Jones’ wealth of love motivates him to go out of his way to help people at great expense to himself. For example, he is not a competent speaker and he suffers from intense social anxiety. But, hoping to convince at least one person, Jones spends many sleepless and miserable nights working on speeches so that at public forums he can argue on behalf starving children. He hopes that they will be given food and shelter and aid in becoming more similar to God. He hardly ever succeeds. There are other examples of this: Jones barely makes enough money to get by. But he often goes without eating because he has given the change that he would have used for lunch to Religious Charity. Unbeknownst to him, however, the money is often misused. It rarely gets to the starving children for whom Jones intended it. While his lack of power prevents his efforts from being very effective, Jones helps people in whatever little ways that he can.
Smith: Smith is enormously powerful and a little loving. While Smith does a lot of good things for people, he would not bother if doing so were any more difficult for him. For example, Smith is a talented speaker. If he is at a coffee shop and people are having a discussion about whether or not it is important to help starving children, Smith will often step in and, in a matter of seconds, effortlessly convince them to give generously. He is always successful and people are changed forever after these brief coffee shop encounters. If it were any harder to do, however, Smith would have no hesitation about stopping. There are other examples of this: Smith makes a lot of money. If a Religious Charity representative comes to Smith’s door asking for a donation, Smith writes them a big check just to get them out of his hair so that he can get back to his main passion—listening to electronic music and drinking beer. As it so happens, Smith is by far Religious Charity’s biggest donor. The money is always used effectively—securing for starving children food and shelter and aid in becoming more similar to God. While he is slightly pleased by this, Smith is mostly just happy that he can continue to wallow in his electronic music and beer drinking.
Max: Max is very loving and very powerful. He makes a lot of money. He gives most of it to the poor. Although he very much enjoys drinking beer and listening to electronic music, he would never think of spending his money or time on those things. He always devotes his time and money to the poor.
Min: Min is only a little loving and only a little powerful. He makes hardly any money. He never gives any of it away. He squanders what little he has on beer and electronic music.
GOOD*, when supplemented with the right way of construing distortion, exaggeration, and unbalance, seems to deliver correct, or at least acceptable, judgments about the comparative intrinsic values of Hitler, Jones, Smith, Max, and Min. Consider Jones and Smith. They are both similar to God in the relevant “holistic” sense. To varying extents they are both powerful and loving and so on. But their power and love are distorted and exaggerated and unbalanced. Of course, the way in which Hitler has the relevant properties is much more distorted and exaggerated and unbalanced than the ways in which Jones and Smith have the relevant properties. So Jones and Smith both faithfully resemble God to a greater degree than Hitler and are therefore intrinsically valuable to a greater degree than Hitler. Now contrast Jones and Smith with Max. Max has the relevant properties in a way that is not distorted or exaggerated or unbalanced at all. What is more, he has the relevant properties in greater proportions than Jones and Smith. So Max faithfully resembles God to a greater degree than Jones and Smith. So he is intrinsically valuable to a greater degree than Jones and Smith. Now contrast Jones and Smith with Min. Min has the relevant properties in a way that is much more balanced than Jones and Smith. But Jones and Smith have at least one of the relevant properties to a much greater degree than Min. The increase that each has in resemblance outweighs the decrease in lack of balance. So Jones and Smith each faithfully resembles God to a greater degree than Min. So Jones and Smith are intrinsically valuable to a greater degree than Min. Finally contrast Jones and Smith. Depending on which account of distortion and exaggeration and unbalance it is combined with, GOOD* will deliver a different judgment about the comparative intrinsic values of Jones and Smith.
4 Against Faithfulness
One odd feature of Finite and Infinite Goods is that Adams never offers an account of overall intrinsic value. He offers an account of intrinsic goodness. He offers an account of various sorts of badness and makes some suggestive comments about intrinsic badness. But he never unifies them in a way that generates an account of overall intrinsic value. Furthermore, it seems to me that the intuitions motivating Adams’ revisions to GOOD are really intuitions about Hitler’s overall intrinsic value rather than Hitler’s intrinsic goodness. As I see it, the intuition is not that Hitler’s power did not contribute to the degree to which he was intrinsically good. Rather, the intuition is that Hitler was so intrinsically bad that whatever intrinsic goodness he had is vastly outweighed by his intrinsic badness. To bring this intuition out, note that GOOD* says nothing about overall intrinsic value. Depending on just how much Hitler’s power, together with his other properties, being distorted and exaggerated and unbalanced is taken to detract from his faithful resemblance to God, GOOD* either delivers the judgment that Hitler was intrinsically good to some positive degree or it delivers the judgment that Hitler was intrinsically good to degree zero. So unless the theory is unified with an account of intrinsic badness, Hitler was either overall intrinsically valuable to some positive degree or he was overall intrinsically valuable to degree zero or the theory delivers no judgment about Hitler’s overall intrinsic value at all. To get Adams’ theory to deliver the judgment that Hitler was overall intrinsically bad, therefore, an account that unifies Adams’ discussion of intrinsic goodness and intrinsic badness is still needed. As I will argue below, there is a simple way to combine GOOD with Adams’ theory of badness that delivers the judgment that Hitler was overall intrinsically bad. Since the unified account of overall intrinsic value is needed anyway and since the unified account, when combined with GOOD, can deliver the desired judgment, the move to GOOD* and to faithful resemblance is unmotivated.
Our discussion has turned up important problems about resemblance—problems in determining whether the relationship between two things constitutes a resemblance and whether an acknowledged resemblance images God “faithfully” enough to constitute an excellence. It appears that both questions may turn on how important the shared properties in the case are.... For a theistic theory it is natural to appeal to God’s view of things or God’s attitude toward things. The theist can appeal to God’s view of things as the definitive standard of importance, and hence of resemblance and of faithfulness, at least where the question is about images of God—saying that considerations in these matters have the importance God sees them as having, and that they resemble, and faithfully image God, insofar as God sees them as doing so.
A number of concerns spring up at this point. One might worry that such an appeal is circular. Indeed, Adams spends a lot of effort arguing that it is not. Or, one might worry that which properties God sees as important, and which ways in which God has his properties are seen by Him as important, could have been very different than they actually are. So although, given what God actually sees as important, Hitler’s power and other properties are had in a way that is distorted and exaggerated and unbalanced, if God had seen importance differently, Hitler’s power and other properties would not be distorted or exaggerated or unbalanced in the relevant sense. So GOOD*’s judgment about Hitler would match GOOD’s judgment about Hitler. So, if Adams reasons for disapproving of GOOD are sound, then GOOD* is hardly an improvement. Of course, there are maneuvers Adams can make to try to avoid this problem. Maybe it is an essential property of God to see importance in the way that He does. Maybe Adams can show that positing this essential property does not lead to circularity. My only point is that all of these complications can be avoided by modifying Adams’ theory in the way that I suggest below. My version of Adams’ theory is compatible with, but does not need to posit, such an essential property. That is a reason to prefer my version of Adams’ theory over GOOD*.
5 The Refined Theory
GOOD: An object, o, is fundamentally intrinsically good to degree, n, if and only if o is similar to God to degree n.
In Finite and Infinite Goods, Adams discusses several kinds of badness and analyzes each in terms of similarity to God.4 It is possible to analyze overall intrinsic value in terms of both fundamental intrinsic goodness and these kinds of badness.
BADI: An object, o, is instrumentally bad to degree, n, if and only if n is the number that is obtained by taking every object that is such that o causes a decrease in the degree to which that object is fundamentally intrinsically good and then adding each of those objects’ decreases.
BADM: A subject, s, is motivationally bad if and only if s is against fundamental intrinsic goodness.
BAD: A subject, s, is intrinsically bad to degree, n, if and only if (i) s is motivationally bad and (ii) s is instrumentally bad to degree n.
GOODO: An object, o, is overall intrinsically valuable to degree, n, if and only if n is the number that is obtained by subtracting the degree to which o is intrinsically bad from the degree to which o is fundamentally intrinsically good.
6 Return to the Hitler Objection
The theory will require further refinements. But, for expository purposes, it is useful to note that the theory, presently refined, delivers the correct judgment about the Hitler objection.
7 The “Hitler was Bad Overall” Reply
The first thing to do, in giving a proper response to the Hitler objection, is to change the desideratum. Adams seems to want the judgment that Hitler’s power did not contribute to the degree to which he was fundamentally intrinsically good. But neither Adams’ theory nor the refined theory can deliver that judgment. As I argued earlier, however, that is not the interesting issue. It is more important to make sure that GOODO delivers the judgment that Hitler was overall intrinsically bad to a very large degree. This can be done without denying that Hitler’s power contributed to the degree to which he was fundamentally intrinsically good.
Here is a way that one might try to show that Hitler was overall intrinsically bad to a large degree: There were many people that Hitler hated who were fundamentally intrinsically good in various respects. So, by BADM, he was motivationally bad. Furthermore, Hitler killed a lot of these people and caused a decrease in the degree to which many of them were similar to God.5 So, by BADI, Hitler was instrumentally bad to a very large degree. So, by BAD, Hitler was intrinsically bad to a very large degree. Now, Hitler was similar to God with respect to some properties. He was very powerful, he was smart, and he was a person. So, by GOOD, powerfulness, smartness, and personhood contributed to the degree to which he was fundamentally intrinsically valuable. But however much these properties contributed to the degree to which he was fundamentally intrinsically valuable, surely, whatever that turns out to be, it will be much smaller than the degree to which Hitler was intrinsically bad. Consider, therefore, the number that is obtained by subtracting the degree to which Hitler was intrinsically bad from the degree to which Hitler was fundamentally intrinsically good. It is a large negative number. So GOODO correctly judges that Hitler was overall intrinsically bad to a large degree.
Although the refined theory is able to deal with this formulation of the Hitler objection, the present refinements are still not adequate. The problem reemerges given a slightly modified statement of the relevant example: There is some number that is the degree to which Hitler was intrinsically bad. Let it be any large number. Now, suppose there is someone, call him Hitler1, who is like Hitler in the following respect: He does only and exactly the bad things that Hitler did. So he is instrumentally bad to the same degree as Hitler. What is more, there are a lot of people that Hitler1 desires to kill and to make less similar to God. So he is motivationally bad as well. So, given BAD, Hitler1 is intrinsically bad to the same degree as Hitler. Next, suppose that Hitler1 is different from Hitler in the following respect: He is much more powerful. Since God is omnipotent, as a person’s power increases, the degree to which that person is similar to God increases along with it.6 So no matter how intrinsically bad Hitler1 is, if he is stipulated to be powerful enough, he will resemble God and, hence, be fundamentally intrinsically good to a much greater degree. So suppose that Hitler1 is powerful enough for this to be the case. Then consider the number that is obtained by subtracting the degree to which Hitler1 is fundamentally intrinsically good from the degree to which he is intrinsically bad. The result is some large positive number. So, if GOODO is true, then Hitler1 is overall intrinsically good. But Hitler1 is not overall intrinsically good. So GOODO is false.
Here is another problem for the refined theory: Consider Hitler2. Hitler2 is identical to Hitler with the exception that Hitler2 is much less powerful than Hitler and any differences entailed by this exception. However, Hitler2 caught many more lucky breaks than Hitler. So, in spite of his comparatively little power, Hitler2 is instrumentally bad to the same degree as Hitler. Now contrast the overall intrinsic value of Hitler with the overall intrinsic value of Hitler2. Other than the difference in power, they are identical in all respects relevant to the determination of their fundamental intrinsic goodness. So Hitler is fundamentally intrinsically good to a significantly greater degree than Hitler2. Furthermore, Hitler and Hitler2 are both motivationally bad and they are instrumentally bad to the same degree. So, by BAD, they are intrinsically bad to the same degree. Now subtract the degree to which Hitler is intrinsically bad from the degree to which he is fundamentally intrinsically good. Do the same for Hitler2. Then contrast the two numbers. The number associated with Hitler will be significantly greater than the number associated with Hitler2. So, if GOODO is true, then Hitler is overall significantly less intrinsically bad than Hitler2. But Hitler is not overall significantly less intrinsically bad than Hitler2. So GOODO is false. The refinements I have made to Adams’ theory are not yet adequate.
8 The Fallen Properties Reply
FALLEN: A property, p, is fallen with respect to a subject, s, if and only if (i) s is motivationally bad, (ii) s is instrumentally bad to degree n, (iii) p is instantiated in s, (iv) p contributes to the degree to which s is similar to God and (v) s’s instrumental badness depends, in the relevant sense, on s’s instantiating p.
(v1) if s had not instantiated p, then the degree to which s is instrumentally bad would have been less than n.
(v2) s’s instantiating p is causally relevant to s’s being instrumentally bad.
(v3) p is one of the properties on which s’s instrumental badness supervenes.7
(v4) s’s instantiating p grounds s’s being instrumentally bad.8
BAD2: A subject, s, is intrinsically bad to degree, n, if and only if (i) s is motivationally bad and (ii) n is the number that is obtained by adding the degree to which s is instrumentally bad to the contributions made to the degree to which s is similar to God by all of s’s fallen properties.
The theory, when refined in this way, is able to deal with the variant Hitler objections. Consider all of Hitler1’s fallen properties. Prominent among these is being powerful. For if Hitler1 were not powerful, then he would not have been so instrumentally bad. Not included among Hitler1’s fallen properties is having a weird moustache. So however much being powerful contributes to the degree to which Hitler1 is fundamentally intrinsically good, by BAD2, the same amount is added to the degree to which he is intrinsically bad. So no matter how powerful he is, that power, when all is said and done, does not make a positive contribution to Hitler1’s overall intrinsic value. What is left to do is to subtract Hitler1’s instrumental badness from the contributions made to the degree to which he is similar to God by his non-fallen properties. The result is some big negative number. Modifying the account of intrinsic badness in this way, therefore, enables GOODO to generate the correct result—Hitler1 is overall intrinsically bad to a very large degree.
Now contrast once more Hitler and Hitler2. By the reasoning given in the previous case, neither Hitler’s power nor Hitler2’s power makes a positive contribution to their respective degrees of overall intrinsic value. Furthermore, the difference in power is the only difference between Hitler and Hitler2 that is relevant to determining their intrinsic values. So the refined theory enables GOODO to deliver the correct judgment—Hitler is not significantly less intrinsically bad overall than Hitler2. Neither the Hitler objection nor any of its variants pose a threat to the refined theory.
9 The “What if God Were Different?” Objection
Suppose that God is very different than He actually is with respect to His desires and His behavior. Suppose that God hates academics. Suppose that He kills some of them and makes others of them stupid. Suppose that I also hate academics. Suppose that I also kill some academics and make others stupid. These properties contribute to the degree to which I am similar to God. So, by GOOD, they contribute to the degree to which I am fundamentally intrinsically good. So, given GOODO, these properties make a positive contribution to my overall intrinsic value. But killing academics and making them stupid does not contribute to my overall intrinsic value. So GOODO is false.
This formulation of the objection is not persuasive. The premise ‘If GOODO is true, then these properties make a positive contribution to my overall intrinsic value’ is false. To see this reason as follows: Academics are persons and they are smart. Even though God hates them, they are similar to Him with respect to these properties. So these properties contribute to the degree to which they are fundamentally intrinsically good. So, by BADM, I am motivationally bad. Note also that by destroying some academics and making the others stupid, I am causing a decrease in the degree to which they are similar to God. So given, GOOD, I am causing a decrease in the degree to which they are fundamentally intrinsically good. So, by BADI, these properties contribute to the degree to which I am instrumentally bad.
Now, note that my hatred of academics and my being such that I kill and render stupid academics are each fallen properties. After all, if I had not killed some of them and caused others of them to be stupid, then I would not have been so instrumentally bad. Furthermore, if I had not hated academics, then I would not have killed them and caused them to be stupid. So I would not have been so instrumentally bad. Given FALLEN, then, the relevant properties are fallen properties. So, since they are fallen, by BAD2, the contribution the relevant properties make to the degree to which I am similar to God entails no corresponding contribution to the degree to which I am intrinsically bad. Furthermore, by BAD2, the contribution that the relevant properties make to the degree to which I am instrumentally bad is added to the degree to which I am intrinsically bad. So the increase in similarity to God contributes nothing to my overall intrinsic value and the instrumental and motivational badness make a negative contribution to my overall intrinsic value. So, GOODO delivers the correct judgment: if God hated and killed and stupidified academics and if I were to do those things, then such things would result in a decrease in the degree to which I am overall intrinsically valuable.
So far only one class of the relevant “What if God Were Different?” cases has been considered. What is common to each member of this class is this: There is some property that contributes to the degree to which various subjects are similar to God. God hates those subjects. God causes a decrease in the degree to which those subjects are similar to Him with respect to those properties. I am similar to God with respect to hating those subjects and causing a decrease in the degree to which they are similar to Him. In cases like these, the refined theory delivers the correct result. However, there is another relevant class of cases that cannot be handled so easily. There are “What if God were Different?” cases in which God tortures some subject but does not in any way cause a decrease in the degree to which they are similar to Him. The refined theory, one might reasonably think, delivers the wrong judgment about such cases.
Suppose that God hates academics. Suppose that once a week, just for fun, God causes academics to experience intense, horrific pain. Suppose, in addition, that God never causes a decrease in the degree to which these academics are similar to Him. If memories of such pain would cause the academics to neglect their studies and become less intelligent, then after each episode of pain God erases the memory. For any way in which such episodes of torture might result in a decrease in similarity to God, He takes steps to ensure that no such decrease occurs. So after each episode of pain, the academics’ degrees of similarity to God are left constant. Before each episode of pain, the academics are smart. After the episodes, they are just as smart. Before, each episode each academic is a member of a community. After, they are still members of the same communities.
Now, suppose that I am very different than I actually am. Suppose that I share God’s hatred of academics. Suppose that I find ways to cause them intense pain without producing any decrease in the degree to which they are similar to God. Maybe I build a machine that will allow me to do this. Maybe I am granted powers that will enable me to do this.
The refined theory, it would seem, delivers the wrong judgment about this case. It judges that, not only do the relevant properties contribute to the degree to which I am fundamentally intrinsically good, but they also make a positive contribution to my overall intrinsic value. To see this, reason as follows: My hatred of academics and my acts of causing them intense pain contribute to the degree to which I am similar to God. So, the relevant properties contribute to the degree to which I am fundamentally intrinsically good. Given BADI, and given that I never cause a decrease in the degree to which any academic is fundamentally intrinsically good, my being such that I perform these actions does not contribute to the degree to which I am instrumentally bad. Since the relevant properties do not contribute to the degree to which I am instrumentally bad, condition (v1) of FALLEN is unsatisfied. So my hatred and acts of torture are not fallen properties. So, by BAD2, such properties do not contribute in any way to the degree to which I am intrinsically bad. So, given GOODO, the relevant properties make a positive contribution to my overall intrinsic value. This is the wrong judgment.
Two replies are available on behalf of the refined theory. First, the premise of the above argument that reads ‘my being such that I hate academics and torture them does not contribute to the degree to which I am instrumentally bad’ is false. Consider again the motivation for this premise. The motivation was that any acts of torture would be erased from the memories of those who were tortured. So no psychological harm that could constitute or counterfactually determine a decrease in the academic’s degree of similarity to God would obtain.
This motivation is inadequate. The academics in question do become less similar to God. Consider one of the academics the day after he was tortured. He reflects on the day before. He considers the proposition ‘I was not tortured yesterday’. He believes that proposition. But his belief is false. So whereas before the academic had n false beliefs, now he has (at least) n + 1 false beliefs. So the academic is further away from having zero false beliefs than he was before. So since God has zero false beliefs and the academic has at least one more false belief than he did a few days ago, the academic is less similar to God than he used to be. And, of course, one does not have to actively consider a proposition in order to believe it. This result generalizes, therefore, to the other tortured academics who never actively consider the relevant proposition. So if I had not hated those academics, then I would not have tortured them. And if I had not tortured them, then they would not have the extra false beliefs. So if I had been without these properties, then I would not have been as instrumentally bad. So, by FALLEN, my being such that I hate and torture academics are fallen properties. Given GOODO, therefore, the relevant properties do not make a positive contribution to my overall intrinsic value. Indeed, the relevant properties make a negative contribution to my overall intrinsic value. So the refined theory can overcome even the present, seemingly formidable, variant of the “What if God Were Different?” objection.
But leave that to the side. Let us grant, for the sake of argument, that there is some way for God and me to torture academics without causing any decrease in the degree to which they are similar to Him and, hence, without causing a decrease in the degree to which they are fundamentally intrinsically good. This leads to the second reply: In order to obtain the desired result, BADI may be revised to include other sorts of badness besides decreases in fundamental intrinsic value. In Finite and Infinite Goods, Adams offers an account of welfare. He analyzes welfare in terms of pleasure, pain, and (what I am calling) fundamental intrinsic goodness.10 A natural revision of the theory, then, would incorporate Adams’ discussion of welfare into the account of instrumental badness. This could be done by reformulating BADI so that it includes not just decreases in a subject’s fundamental intrinsic goodness but also decreases in a subject’s welfare. Then my being such that I hate and torture academics would contribute to the degree to which I am instrumentally bad. So condition (v1) of FALLEN would be satisfied. So, by BAD2, those properties would make a negative contribution to the degree to which I am overall intrinsically valuable.11
10 The “What if Adams Believed he was God?” Objection
Suppose that Adams believed he was God.12 If he were to believe this, then the degree to which he is similar to God would be increased. So, given GOOD, the degree to which Adams is fundamentally intrinsically good would be increased. And, given GOODO and BAD2, if Adams were to believe he is God, then the degree to which he is overall intrinsically valuable would be increased. But the degree to which Adams is overall intrinsically valuable would not be increased if he were to believe that he is God. So, GOODO is false.
This formulation of the objection is not persuasive. The premise ‘If Adams were to believe he is God, then the degree to which he is similar to God would be increased’ is false. To see this, consider David Lewis’ (1973, 1979) analysis of counterfactuals.13 Let @ be the actual world. Then contrast two possible worlds W1 and W2.
W1: Adams believes ‘I am God’ at W1. The behavior of Adams at W1 does not differ in any way from the behavior of Adams at @. At both worlds he is a philosopher, he is the author of Finite and Infinite Goods, he goes to church and receives the Eucharist, he has exactly the same conversations and friends, and so on. The beliefs of Adams at W1 are almost identical to the beliefs of Adams at @. The only difference between the two is that Adams at W1 has the extra belief that ‘I am God’. There is no belief that Adams at W1 has that Adams at @ does not have. For every belief that Adams at @ has, including ‘Adams is not God’, Adams at W1 also has it.
W2: Adams believes ‘I am God’ at W2. The behavior of Adams at W2 is very different from the behavior of Adams at @. Instead of going into philosophy, Adams at W2 joined a cult. That is how he came to believe he is God. His personality is very different. He has long hair and he frequently enjoys marijuana cigarettes. He does not have very many of the same friends that he has at @. He does not do any serious philosophy or go to church. The beliefs of Adams at W2 are very different from the beliefs of Adams at @. Adams at W2 has a lot of crazy and cultish false beliefs. He tries to fit together his belief that he is God with his other beliefs. Adams at W2 has many more false beliefs, such as ‘Adams is God’ and far fewer true beliefs, such as ‘Adams is not God’, than Adams at @.
Given Lewis’ criteria for determining closeness between possible worlds, W2 is closer to @ than W1.14 So the relevant counterfactual must be evaluated at a W2-like world rather than a W1-like world. But Adams is less similar to God at W2-like worlds than he is at @.15 So the relevant counterfactual is false. So this formulation of the objection is unpersuasive.
11 A Different Formulation of the “What if Adams Believed he was God?” Objection
There is another, more plausible, way of formulating the objection. The real worry is not about whether the counterfactual in question is true or false. The worry is instead that there is some W1-like possible world, no matter how remote from @ with respect to psychological laws, at which Adams believes ‘I am God’ and everything else is left the same as it is at @. At W1 the degree to which Adams is similar to God is greater than it is at @. So if GOODO is true, then the degree to which Adams is overall intrinsically valuable at W1 is greater than the degree to which Adams is overall intrinsically valuable at @. But the degree to which Adams is overall intrinsically valuable at W1 is not greater than the degree to which Adams is overall intrinsically valuable at @. So GOODO is false.
While this formulation of the objection is more persuasive than the first, it is unsound. By believing ‘I am God’, Adams at W1 gains a belief in common with God. But its truth-value is relative to the person who believes it. The belief is false with respect to Adams and it is true with respect to God. So, while Adams at W1 shares a belief with God that Adams at @ does not, Adams at @ is more similar to God than Adams at W1 with respect to the avoidance of false beliefs. This property that Adams at @ keeps and Adams at W1 loses has more weight in determining similarity to God than the belief that Adams at W1 gains in common with God does. So Adams at W1 is not more similar to God than Adams at @.
The thing to do is not to start by deciding, once and for all, what we think about similarity of worlds, so that we can afterwards use these decisions to test [my analysis]. What that would test would be the combination of [my analysis] with a foolish denial of the shiftiness of similarity. Rather, we must use what we know about the truth and falsity of counterfactuals to see if we can find some sort of similarity relation—not necessarily the first one that springs to mind—that combines with [my analysis] to yield the proper truth conditions. It is this combination that can be tested against our knowledge of counterfactuals, not [my analysis] by itself. In looking for a combination that will stand up to the test, we must use what we know about counterfactuals to find out about the appropriate similarity relation—not the other way around.
Given how influential Lewis’ analysis is, I think that I can help myself to the same strategy when testing Adams’ theory of intrinsic value. Let us use what we know about intrinsic value to discover the appropriate similarity relation to God. What we find, if we do that, is the similarity relation that assigns great importance to increasing one’s match to God with respect to the avoidance of false beliefs and assigns little or no importance to increasing one’s match to Him with respect to His de se beliefs such as ‘I am God’. Combined with Adams’ theory this similarity relation delivers the correct judgment: Adams at W1, that is Adams at a world at which he believes ‘I am God’ but everything else, other than the psychological laws, is left the same as it is at @, is not more similar to God than Adams at @. So the “What if Adams believed he was God?” objection is unsound.
When modified in appropriate ways, Adams’ theory is able to deal with a number of natural, seemingly formidable, objections. Adams’ Finite and Infinite Goods, I conclude, is suggestive of an interesting theory of intrinsic value that is worthy of critical attention.
Finite and Infinite Goods is the subject of a symposium in Philosophy and Phenomenological Research. A Theory of Virtue is the subject of a symposium in Philosophical Studies (Adams 2006).
Authors, for example, as diverse as Richard Boyd, Fred Feldman, and Mark Johnston have praised various portions of Adams book. Boyd (2003) says that “Finite and Infinite Goods makes an especially important contribution to our understanding of the naturalist realist project in metaethics.” Feldman (2004, p. 17) calls Adams’ discussion of welfare in Finite and Infinite Goods “impressive.” Johnston (2009, p. 25) says that Finite and Infinite Goods is “magisterial.”
The accounts of instrumental badness and motivational badness are not necessarily meant to be accounts of our ordinary concepts of instrumental and motivational badness or anything like that. They are instead technical terms introduced to enable the theory to avoid the three objections that concern me in this paper.
I should illustrate this point with some examples: Hitler subjected some Jews to experiments and withheld from them proper nutrition. These practices damaged their cognitive faculties and abilities. That is a decrease in the degree to which they are similar to God. Before the Holocaust there were many Jews who were members of communities. Hitler broke many of these communities up. Since God is a member of a communities (the Trinity and His people) that is a decrease in the degree to which they were similar to God. Hitler caused many Germans to hate Jews. Since God loves Jews that is a decrease in the degree to which those Germans were similar to God.
This premise is questionable. But grant it for the sake of argument.
Thanks to Fred Feldman for suggesting (v3).
I have heard two complaints about modifying Adams’ theory in this way. First, I have been told that the modification I introduce is “ad hoc” and that I am “gerrymandering” Adams’ theory. Perhaps what someone who offers this objection means is that I introduce the fallen properties revision without any motivation other than the fact that it makes Adams’ theory produce the correct judgment about Hitler and Hitler1 and Hitler2. But, such people seem to be claiming, modifications to a theory need independent motivation. So the fallen properties reply is defective.
Second, I have been told that I rely on a confusion about the relation between intrinsic and extrinsic value. My account of fallen properties implies that the intrinsic value of an object is sometimes altered by the extrinsic value of that object. But, as Moore has taught us, such a suggestion is incoherent and unintelligible. So my modification of Adams’ theory is defective.
It is important to see that these objections are mistaken. Concerning the complaint about ad-hoc-ness and gerrymandering: The objector is holding me to an unreasonably high standard. Consider Feldman (1983). In some passages he introduces modifications to utilitarianism with no other motivation than the fact that those modifications make utilitarianism deliver the correct judgment about a class of cases. Sider (1991, 1993) does the same thing. This is a common strategy. So the way in which I develop Adams’ theory is in accordance with tradition.
I will, however, leave all of this to the side. I can meet the objector’s unreasonably high standards. There is a precedent, in the literature, for the sort of move I have made. Shelly Kagan (1998, p.281) has argued, on the basis of several examples, that the extrinsic value of an object can alter its intrinsic value. He says this: “I want to leave open the possibility that the intrinsic value of an object may be based (in part) on its instrumental value.” Much more recently, while commenting on Adams’ theory, Kagan (2009a, p. 390) repeats this line. Josh Parsons (manuscript, p. 1) takes a similar line. He says this: “Suffice it to say that... objects can sometimes be instrumentally intrinsically valuable, and sometimes be finally extrinsically valuable... and let it be taken that people sometimes use “intrinsically valuable” when they clearly mean “valuable as-an-end”, and use that misleading phrase to conflate the two.” So there is a precedent in the literature, based on counterexamples to orthodoxy, for the sort of move I have made. Therefore, although I do not need it, there is independent motivation for my introduction of fallen properties.
Mooreans do indeed deny that for any x, the uniqueness or usefulness of x is relevant to x’s intrinsic value. But I know of no evidence that Mooreans would deny that for any x, states of affairs ascribing uniqueness or usefulness to x could have intrinsic value. In fact, it must be pointed out that many philosophers within the Moorean tradition have thought that states of affairs ascribing non-intrinsic properties have intrinsic value.
In light of Bradley’s work on this topic, the fallen properties reply, could easily be translated into terms that a Moorean would find unobjectionable. Following Bradley’s strategy, one could recast Adams’ theory in terms of states of affairs rather than objects. So a state of affairs ascribing fallen properties to an object would have less intrinsic value than one that did not. The relevant issue, therefore, is whether objects or states of affairs are the primary bearers of intrinsic value. Adams’ theory can be formulated so as to accommodate either outcome.
Keep in mind here that fundamental intrinsic value and overall intrinsic value are distinct.
Several people have presented me with one variant or another of the following objection: “God is infinite and everything else is finite. So nothing is more similar to Him than anything else. Consider the case of beliefs. Suppose that I have 5 true beliefs and you have a billion. In that case we are both equally far from God’s omniscience. So, neither of us is more similar to God than the other with respect to being omniscient.” It is important to see that this objection is mistaken. Consider similarity to God with respect to the avoidance of false beliefs. God has 0 false beliefs. Suppose that I have 40 billion false beliefs and that you have 13 false beliefs. 13 is much closer to 0 than 40 billion. So you are much more similar to God with respect to the avoidance of false beliefs than I am. Consider similarity to God with respect to the probability that one’s beliefs are true. The probability that God’s beliefs are true is always 1. The probability that our beliefs are true varies. But suppose that on this occasion the probability that my beliefs are true is .2 and yours is .75. A probability of .75 is much closer to 1 than a probability of .2. So you are much more similar to God with respect to the probability that one’s beliefs are true than I am. Consider being similar to God with respect to, not the number of beliefs He has, but instead each particular belief that He has. Suppose I believe the proposition ‘2 + 2 = 4’. I am not more similar to God than my counterpart who does not believe this proposition with respect to having the same number of beliefs that He does. This is for the very reason that the objector points out. But I am more similar to God, than my counterpart, in this sense: God and I share the property of having the belief ‘2 + 2 = 4’. Aside from beliefs: Consider the property of being a person. God has the property of being a person. You and I have the property of being a person. You and I sharing this property with God makes us more similar to Him than, say a rock or a book, is. Consider being actualized or existing. God exists and is actual. You and I exist and are actualized. So in this respect we are more similar to Him than some merely possible or non-existent object.
Lewis’ analysis is this:
A counterfactual ‘If it were the case that A, then it would be the case that C’ is (non-vacuously) true at a possible world, W, if and only if there exists some (accessible) possible world at which A and C are true that is closer to W than any possible world at which A is true and C is false.
It is of the first importance to avoid big, widespread, diverse violations of law.
It is of the second importance to maximize the spatio-temporal region throughout which perfect match of particular fact prevails.
It is of the third importance to avoid even small, localized, simple violations of law.
It is of little or no importance to secure approximate similarity of particular fact, even in matters that concern us greatly.
To see this reason as follows: According to Lewis, avoidance of a big widespread and diverse violation of @’s laws has more weight in determining closeness to @ than matching @ with respect to particular matter of fact. Furthermore, Adams’ belief that ‘I am God’ at W1 requires such a violation of the psychological laws governing beliefs with respect to @. Here is why: No one at @ has beliefs that are so dramatically causally isolated from each other and from behavior in the way that Adams’ belief at W1 is. Adams’ belief at W1 was not caused by anything. For each instant of time at which Adams believes ‘I am God’ at W1, he is held back from forming closely related beliefs such as ‘Adams is God’ and from dropping beliefs that are obviously inconsistent with ‘I am God’ such as ‘Adams is not God’. Adams’ belief never shows up in his behavior. Not even his psychotherapist can tell that Adams believes ‘I am God’. Not even Adams believes ‘Adams believes he is God.’ At W2, on the other hand, Adams’ belief conforms to the psychological laws of @. His belief is caused by a thorough brainwashing combined with a series of ecstatic marijuana induced rituals at the cult compound. His belief causally interacts with his other beliefs producing and being reinforced by them. His belief has a causal influence on his behavior. Though the result is much less of a match with @ with respect to particular matter of fact, it avoids a violation of the psychological laws of @. So W2 is closer to @ than W1.
To see this, reason as follows: Adams at W2 shares one new belief with God, namely ‘I am God’, that Adams at @ does not have. But Adams at W2 gains many crazy and false beliefs that God does not have such as ‘Adams is God’ and ‘God should leave the compound sometime today and head over to the grocery store to get some stuff for the fridge’ and he loses many sane and true beliefs that he has in common with God at @ such as ‘Adams is not God’ and ‘God was not born in the 20th Century’. So at W2 Adams is less similar to God than he is at @.
Thanks to Lynne Baker, Andrew Dole, Pete Graham, Gary Mathews, James Patten, Alex Sarch, the members of Lynne and Gary’s Fall 2008 Philosophy of Religion Seminar, and the two reviewers for Ethical Theory and Moral Practice. Special thanks to Lowell Friesen for many insightful discussions of this topic over coffee. Most of all I thank Fred Feldman for extensive commentary and discussion of countless drafts of this paper.