Questioning Fyfe’s Desirism, II

In the thread that followed the Introduction, we discussed,

• The type-token distinction as it applies to desires;
• The DFAU/DU distinction;
• The concept of strength as it applies to desires;
• Thomas Reid's objections 1 and 3, and faithlessgod's replies (indexed in Desirism Notes);
• Peripheral issues such as DCT, my use of useful, etc.

I'd say more than enough has transpired to warrant a second post, and I'd like to discuss:

1. How the discussion with Thomas Reid, TaiChi and faithlessgod has impacted my position;
2. Why my primary objection to desirism remains;
3. The hierarchy of desires concept;
4. Old questions that remain unanswered;
5. New questions that resulted from our discussion.

1) How did our discussion impact my position?

Although our discussion was fairly involved, I didn't gain much in terms of new understanding or answered questions. I had opened by expressing my general sentiments on desirism:

1) I agree desirism is efficient in that it provides a working formula for determining the most useful course of action relative to some desire,

2) I agree desirism is accurate in that it accurately describes how individuals tend to make moral decisions in the real world, but

3) neither desirism's efficiency nor its accuracy necessarily make it true, or even worthy of being called the best moral theory.

I still hold to 1 and 2 exactly as stated, and I'd also like to emphasize that useful in 1 should be interpreted in the everyday sense of the word. It is not meant to correlate to any tenet of desirism such that it should be judged accordingly. Regarding 3, I've realized the need to clarify: the "desirism equation" is true, mathematically, meaning that for any given agent(s), any particular desire(s) will tend to either fulfill or thwart other desires, and whenever we know the range of relevant desires, this relationship is always objectively quantifiable. Further, the relationship is not necessarily either/or, as in many cases, any particular desire(s) will tend to both fulfill and thwart other desires, where "other desires" refers to the balance of all other [affected] desires that exist amongst the agent(s). The manner in which desires relate to other desires is the issue here.

In Desirism Notes I argued that this relationship can be represented by a mathematical ratio:

..say that we have an agent with some desire X. The balance of desires can be represented by a ratio (i.e. 37/48). The first integer represents the total number of desires that would [tend to] be fulfilled if the agent realizes a state of affairs where [desire] X has been made true. The second integer represents the total number of desires that would [tend to] be thwarted if the agent realizes a state of affairs where [desire] X has been kept untrue.

In this example, 37 other desires would be fulfilled if the agent realizes X, while 48 other desires would be thwarted if the agent realizes X. According to desirism, then, X in this case represents an evil desire, because its overall effect on the balance of desires is to thwart more desires than it fulfills. Desirism doesn't evaluate what one, two or all people believe about [desire] X; it evaluates the effect of [desire X] on all other [affected] desires – the balance of desires. The ratio remains [objectively true] regardless of whether we believe desire X will fulfill or thwart other desires.

At this point I need to draw a distinction. When I say that the "desirism equation is true," I do not necessarily mean to imply that a faithful application of desirism will always lead humans to an answer that agrees with their moral intuitions. In actuality, true is a potentially confusing term here, because desirism's strength that I wish to highlight is simply the fact that at any given time, the relationship between any given desire(s) and the balance of desires is objective and quantifiable. Accordingly, any moral question could theoretically be answered objectively and quantified numerically, i.e. in terms of how many other desires it tends to fulfill vs. how many other desires it tends to thwart. This is what I mean when I refer to the fact of desirism's objectivity. and to help make things more clear, I'll refer to this ratio as the desirism equation in the future.

If I were to restate my general sentiments to reflect this acknowledgment of desirism's objectivity, it would go something like this:

1) I agree desirism is efficient in that it provides a working formula for determining the most useful course of action relative to some desire,

2) I agree desirism is accurate in that it accurately describes how individuals tend to make moral decisions in the real world,

3) I agree desirism is objective in that at any given time, the relationship between any given desire(s) and all other desires is objectively quantifiable, but

4) these strengths do not make desirism worthy of being called the best moral theory, IMO.

2) Why does my primary objection to desirism remain?

My primary objection to desirism is quite simple: to say that desire X tends to fulfill more than thwart other desires is insufficient grounds for calling desire X 'good' (in the sense that everyday people tend to use the word). If the [balance of] 'other desires' are not what we'd call 'good' to begin with, then a desire cannot be called 'good' simply because it tends to fulfill more than thwart them. Numerical comparisons of desires tell us nothing reliable about their moral worth. To date, I've not seen a persuasive argument from any desirist suggesting otherwise. So, my objections remain.

3) What do I mean by the phrase hierarchy of desires?

One could visualize the concept by imagining a tiered-pyramid, much like the one depicted on the reverse side of the dollar bill. The apex of the pyramid (the all-seeing eye) represents the agent's desire-as-ultimate-ends. If that phrase sounds confusing, don't worry: we have several examples of real-world language we use to illustrate this concept. The phrases "top priority", "life's work" and "goal in life" all refer to an agent's desire-as-ultimate-ends. An agent's desire-as-ultimate-ends might also be comprised of multiple desires, by the way; an agent's desire-as-ultimate-ends need not be a single desire in every instance.

Lower tiers represent the agent's sub-desires, which may or may not tend towards the direct fulfillment of the agent's desire-as-ultimate-ends, but certainly should never thwart the direct fulfillment of the agent's desire-as-ultimate-ends, else, we have an example of what Fyfe (and myself) would call an irrational desire: not irrational in the sense of lacking evidence, but irrational in the sense of working against the agent's desire-as-ultimate-ends.

One advantage of this concept is that it allows us to numerically quantify an agent's (or set of agents') unique balance of desires, such that we can use these numbers in meaningful, mathematical evaluations. So hold that thought; at this point, I'm simply introducing this concept, not necessarily making any steadfast claims about its implications – yet.

4) Which old questions remain unanswered?

Honestly, so many that I've literally lost track of them all. The first one that comes to mind is what the desirist should do when the numbers are at odds with our moral intuitions. As is the case with Cartesian's Nazi example, or my own Canaanite example – both of which are iterations of the 1000 sadists problem – what happens when a desire undeniably tends to fulfill more than thwart all other [affected] desires, yet our moral intuition strongly suggests that the desire is evil? To show that I'm not merely dismissing the desirist argument, let's take a quick look at Fyfe's attempted defense against 1000 sadist objections. I say "attempted" because it fails right here:

In any society, the more prevalent and the stronger this desire [to torture children] becomes, the more other desires are thwarted. (Fyfe, brackets mine)

No, in any society with pre-existing reasons for action to promote aversion to the desire to torture children, THEN "the more other desires are thwarted." Here, Fyfe loads the evaluation with agents who have pre-existing reason to promote aversion to the desire in question. Remember: since there is no intrinsic value in Fyfe's desirism, then any desire can be 'good' so long as it fulfills more than thwarts other desires. Accordingly, any society that has reasons for action to promote child torture – such that the desire would tend to fulfill more than thwart other desires – has reason to "turn the knob up" for that desire.

We don't need to retreat to unlikely hypotheticals to illustrate this concept. We can refer to historical cultures like the Canaanites, who sacrificed their children to Molech and "passed them through the fire." Of course I will grant that their children's desires were always thwarted, and in that sense, turning the knob up always thwarts some desires by default. However, we are to evaluate the balance of all [affected] desires, not just the children's. The more one turned this knob up in Canaanite society, the more other [affected] desires were generally fulfilled. This is because attaining the favor of the Baalim and Baalot was [arguably] the Canaanites' desire-as-ultimate-ends. In their culture, an entire array of sub-desires were believed to be satisfied if they attained this favor. Agriculture was thought to flourish, as just one example. Though it might make us sick to our stomachs, to the Canaanites, these were reasons for action and the desire to pass children "through the fire" tended to fulfill more than thwart all other [affected] desires. Besides, it's not like children are a limited resource; they could always create more children anyways. If our ONLY criteria for 'good' is 'tends to fulfill more than thwart other desires,' then it's hard for me to see how this wouldn't be genuinely classifiable as a 'good' desire. Anyone?

Now, note that at best, this is only problematic for desirism's definition of 'good' because Fyfe's desirism would never accept "I am gaining the favor of the Baalom and Baalot" as a proposition that could be made true. Remember, Fyfe's desirism discards intrinsic value, the social contract, categorical imperatives, the decrees of the gods, etc. Although the Canaanite example is certainly relevant in that it illustrates the problem with defining 'good' as 'such as to fulfill other desires,' Fyfe need not even take the time to evaluate it because he can simply dismiss the Canaanites' reasons for action as insufficient. So, we'll need an even better example, one where the reasons for action cannot be dismissed as insufficient, but let's save that for another post as this one's already grown too long. Here, I simply mean to say that Fyfe's attempted 1000 sadists defense fails because we have to evaluate the [likely] effects of desire X against all [affected] desires of the agent(s) in question.

5) What new questions arose as a result of our discussion?

Again, too many for me to list, but here's the first one that comes to mind: desirism's defenders say we are to evaluate desires, not acts. Okay, but since desires don't tend fulfill or thwart other desires unless acted on, what do we gain by evaluating desires as opposed to acts?