On Orders of Desire & Value Pluralism

In reaction to the following question posed:

Suppose I have a 1st order desire A; a 2nd order desire B which is to not have desire A; and a desire C which is to not have conflicting desires. Is C a 1st order, 2nd order or 3rd order desire?

It can be achieved by changing desire A to conform with B (and if I wanted exactly that then it would have been a 2nd order desire) or by changing desire B to conform with A (and if I wanted exactly that then it would have been a 3rd order desire), but what if I don’t have any preference as to how to achieve C?

When one speaks of “orders” of desire, it is likely one is looking at the issue propositionally, as the idea of “orders” is a logical one, hence one that applies to propositions. Namely: (1) “I desire 𝓍,” (2) “I desire that it not be the case [I desire 𝓍],” etc. In which case, if there is some desire for the consistency amongst desires, such that one can conceive of the desire as higher-order, such that it becomes “I desire that it not be the case that [I desire that it not be the case [I desire 𝓍]],” what occurs is merely further inconsistency in evaluating the truth or falsehood of any particular proposition about desire, as stating a desire does not seem to mean simply stating a factual state of affairs, but stating something about a factual relation to some event or object that may well be non-factual–and if one were intent on further solving for that evaluative undecidability (that, in a determinate form, would result in inconsistency) by searching for some foundational desire, one gets into an infinite regress insofar as the number of possible orders of desire are indeterminate. And in any case, if that foundation were found or the regress arbitrarily limited, it would deviate from our actual diversifying, everyday, concrete desires. Yet if that arbitrary limit is not reflective of the nature of simple desire as such, such that the limit is actually a priori given as “I desire or do not desire X” (i.e., a point of indecision rather than a real limit), it is possible that such an arbitrary limit itself remains the consequence of at least some second-order desire, such that desires are at least limited up to the second order for purposes of description (that is, in order to be put them in propositional form).

But in the case of the second-order limit desire-proposition, one assumes a contingent decision has to have already been made, and, insofar as the possible choices are constrained in their occurrence according to at least this factor of desire (the factor we are isolating here, the second-order desire), that means the arbitrary limit can only be β€œjustified” or caused precisely if a regress was presupposed. Proposing an arbitrary limit or foundation for the sake of allowing assessment of the truth-value of desire-propositions, or being able to make propositions about desire, would at the same time mean excluding the issue of decision-making which is either a condition of possibility for desire or an outcome of desire. This means the regress problem applies not only in the case of negative desiresΒΉ that are more than or equal to the second order of propositions, but to positive desiresΒ² at any order as well. But this looks like an infinite regress merely from one perspective–if the several orders of desire is herein being treated as a hierarchy, than there is no reason to assume there’s anything bidirectional relationship occuring in this context.

All these objections would ultimately come down to the desire for consistency or inconsistency being located outside of the given orders of desire, in the same sense that in a formal system there will ultimately be a presented undecidability between consistency and inconsistency when striving for soundness. Yet it’s not even all that clear there really is a consistency problem in this case despite the analogical structure between orders of desire and orders of justification–the objects of the proposition “I desire that it not be the case that [I desire 𝓍]” and “I desire 𝓍,” or “I desire that it be the case that [I desire 𝓍]” and “I desire ¬𝓍”Β³ are different–only one of the orders in the second-order/first-order pairs provided here is about 𝓍. The idea that there is an inconsistency presupposes that desiring that it not be the case one desire 𝓍 entails something about what happens to 𝓍 as much as the desire for 𝓍 does. That is, it implies that all desires with the same object are functionally equivalent.

To the contrary–the former second-order desire regards a desire and not the object of that same desire, to that effect producing or suggesting paths of action that would differ radically from a simple negative desire towards 𝓍. In other words, one is presupposing a set of conditionals of the sort: (1.a.) “If there is an 𝓃th-order positive desire regarding 𝓍, then 𝓍 will/shall be the case,” (1.b.) “If there is an 𝓃th-order negative desire regarding 𝓍, then ¬𝓍 will/shall be the case.” If these conditionals are rejected, there is no inconsistency to solve, and so there is no desire for consistency or inconsistency of desires possibly implicated. That is, the perceived inconsistency, circularity, or regress comes from too straight-forward a view of the relationship between desire and action, and thereby too naive an assumption that statements about desire have the same consistency requirements as statements about the world. Desires are rarely, in other words, merely intended towards objects, but implicitly already involve possibilities of action and approach given a combined state of objective and subjective affairs. The regress of orders of desire is an illusory effect of the fact that desire both produces and is produced by this state of affairs. The problem across the orders of desire isn’t so much one of inconsistency or regress as of practical action: its indeterminate what will be the case, or what will be desired in the future and, consequently, what can and should be desired now–it is here that the conflict plays out. This is why desires are nonetheless experienced as conflicting–desire is unitary and singular as a vague generality without a determinate object, but it splinters into agonistic pieces when active reality breaks through the intertia of subjective affect.

headonsilverplate

And in some sense, desires do not just appear to conflict, but desires actually do conflict in the aforementioned latter sense–just not in a logical sense. This is where a set of universal practical dilemmas are derived: scale v. span/scope of decisions (long-term v. short-term, local v. global effects), as well as opposite directionality of effect internal to action itself (anticipation v. production of some outcome, preparation v. follow-through of some process) that produces imperfect information. Imperfect information regarding the erotic or desirous content of future experience, and its relationship to affect at the finalization of an action. Imperfect also in the fact that holding any particular thing as an object of desire automatically segregates it, at a cognitive level, from other phenomena that nonetheless constitutes or captures that object while likely affecting the production or capacity of desire. So the real question is how one’s desire or lack of desire for 𝓍 relates to both 𝓍’s and the desire’s own conditions of possibility. In turn, the question also becomes how these conditions of possibility affect how the relationships among (positive) desires are to be evaluated or how the relationship between a (positive) desire and itself is to be evaluated, insofar as any given (positive) desire could lead to a network of (negative or) opposing desires given the latent potentials of the object of (positive) desire.

The relevant desire is then not of consistency of desire, but of the consistency among actions in upholding the possibility of the fulfillment of any given action–that is, the integrity of action. Integrity is often described as honesty with oneself regarding one’s own moral principles and how one’s acts fit within those principles–but this is rather an epiphenomena of actual integrity, as inconsistency between cognition/conation and action is the bread and butter of all passionate action. That is to say, integrity in this case does not indicate a lack of hypocrisy–if hypocrisy is treated broadly as any observed contradiction in one’s actions (the quintessential being a contradiction between what one declares belief in and what one does). Rather, integrity is a stance towards consistency in relation to the conditions of existence for the desired object and the capacity to affect that object, as well as a stance towards those conditions necessary to reasonable reproduction of that desire. Desirous or erotic action therefore inherently involves a working through the inconsistencies of practical life, which will mean starting from these inconsistencies as bias and moving towards some attempt at greater consistency. It also otherwise involves existing between temporal antinomies (past v. future).

The “orders of desire” are really a helpful but limited gimmick when it comes to describing this–the “orders” really just have different objects of desire that reflect a dialectical movement in the development of desire. Going back to our conditionals, they would more accurately be: (2.a.) “if there is an 𝓃th-order positive desire regarding 𝓍, then 𝓃-𝟏st-order positive desire regarding 𝓍 will be the case and 𝓃-𝟏st-order negative desire regarding ¬𝓍 will/shall be the case” which is practically the same as to say “if there is a positive desire regarding 𝓍, then 𝓍 will/shall be the case”; and (2.b.), being an inversion of the former as far as the positivity/negativity of 𝓍 and of the 𝓃th- or 𝓃-𝟏st-order desires. In other words, we can already see that there is a kind of churning wheel of negation that constitutes the orders of desire. Perhaps the problem then is that we are assuming a propositional nature for desires, and thus that desires about desires are “second order” in a propositional sense. Instead, it may well be that desires have a metaphysical contingency to them, but nonetheless a sort of subsequent logic (dialectical rather than propositional) that makes desire look like a pseudo-coherentist affair.

Yet, supposing we wish to maintain the propositional form, perhaps modal propositions need to be introduced. Suppose that propositions about desire equal to or beyond the second order are actually different in form in a sort of patterned way: e.g., rather than a second-order desire taking the form, “I desire that it not be the case that [I desire 𝓍],” it would be more-so “I desire that it be possible that [I desire 𝓍]” or “I desire that it necessarily be the case that [I desire 𝓍].” In this case the situation becomes far more interesting–one could argue that the language of necessity or possibility renders even propositional inconsistency a non-issue and thus erases it as a possible erotic/desirous concern.

Either of the solutions presented here would render locating a “desire for consistency or inconsistency” a non-problem. The result, of course (if we’re using “desire” as the basis of, or a significant complicating factor in, our ethics or axiology), is a sort of value pluralism–but the role of practical and material conditions (and their various levels/scales) can constrain or expand its scope, such that some value dilemmas indeed are rationally decidable and are thus non-arbitrary despite the intrinsic contingency of desire. That is to say, under an ethics developed from the point of view of desire, traditional kinds of moral rationalism would be scrapped. A new moral rationalism that accepts the indeterminacy and contingency of the ends of action would be birthed, whose concern would be internal instrumental consistency within action and among actions at any given scale. Preference of how to achieve π’ž (as specified in the question) may be a matter of indifference to desiring subjects, but what was just explained suggests that the preservation, expansion, degradation or constraining of possibility is itself going to have ethical/axiological implications that slips past any sort of ethical consequentialism. Ethical consequentialism would play a role, but would be perceived as a severely limited approach due to its negligence of internal instrumental consistency. Neglecting such comes at a high cost in ethics, because it makes it more difficult to find a link between personal and sociopolitical morality, and makes it less likely that ethical teaching could have any observably positive effect on social life. In this regard, an ethics informed by desire as it is conceived in this post could be called a kind of “ethical structuralism” insofar as conditions of possibility are an object of concern for all desires and all action, even if the connection is not immediately apparent for subjects (hence the possible indifference to the means of achieving π’ž).


  1. Negative desires are desires not to do something.
  2. Positive desires are desires to do something.
  3. While this was left out as being a tangential concern, it is certainly possible that there is a material difference between “I do not desire 𝓍” and “I desire ¬𝓍”–that is, between a negation of desire and desire for the negative of some object. For example, if I say I do not desire coffee right now, this is to say that this object is uninvolved in any of my desires. Not only do I not desire coffee, but I consequently do not desire not coffee. That is, in the sense that one could not conclude that I desire everything or anything else. On the other hand, if I say I desire not coffee, I am indeed saying I desire everything or anything but coffee. Perhaps this is why the grammatical convention in natural language use is to negate the verb or adverb, or implicitly negate the copula.