Managing Temptation: Comments on Chrisoula Andreou’s ‘Micromanagement and Poor Self-Control’

: In ‘Micromanagement and Poor Self-Control’, Chrisoula An-dreou argues that some cases of poor self-control are best understood as arising from poor self-management, in particular a kind of intrapersonal micromanagement. She argues that this furnishes us with a better understanding of those cases than the orthodox foreign force paradigm does (on which poor self-control amounts to diminished self-control). I argue that we cannot do without the foreign force paradigm to explain the cases that Andreou discusses. I suggest a both/and approach on which poor self-management and diminished self-control together explain poor self-control.


I. INTRODUCTION
Confession: I took far too long to write this paper.I certainly planned on writing it in good time.But shortly before agreeing to write it, I purchased Zelda: Breath of the Wild.And the first day that I sat down to write, I figured I could put off writing to play a little Zelda.The same thing happened on the second day.And the third.And (to my shame) so on for an entire month.To be clear, when I step back and compare the outcome 'Play 30 days of Zelda and do not write the paper' with the outcome 'Play a few days of Zelda and write the paper', I unhesitatingly judge that I prefer the latter to the former.
The purely hypothetical case just described is an example of poor selfcontrol.I had a goal-write the paper in good time-and yet failed to achieve it.And this was not the fault of any external impediment or coercion.Rather, I engaged in self-defeating behavior by, somehow, inadequately controlling myself.
In 'Micromanagement and Poor Self-Control' Chrisoula Andreou argues for a novel way of understanding this phenomenon.The classical way of understanding it she dubs the foreign force paradigm: poor selfcontrol involves diminished control of the self because one is controlled by forces that are "part of one's psychology [but] external to the set of motivating forces with which one identifies" (169). 1 By contrast, Andreou argues that at least some instances of poor self-control arise from a failure to appropriately manage the self.In particular, she argues that if I have cyclic preferences, then I might exercise poor self-control due to micromanagement: I make a plan (say to stop playing Zelda after 10 days) but subsequently second-guess that plan enough times that I end up realizing an unacceptable outcome.
I am sympathetic to both cyclic preferences and the idea that we can problematically micromanage ourselves.What I want to explore here is Andreou's claim (171) that thinking in terms of micromanagement provides a better explanation of (some cases of) poor self-control than the foreign force paradigm does.I will suggest, by contrast, that we should appeal to both managerial failings and the foreign force paradigm together to understand such cases.This means that while Andreou might be right that "research aimed at figuring out what it takes for selves that are already in control to qualify as managerial successes" (170) is important, it must be pursued in tandem with the more traditional project of "research aimed at locating and empowering the self"(169).

II. PREFERENCES AND PLANS
Central to this discussion is the idea of preference:  ≻  denotes that option  is preferred to option .You can think of this as saying that the agent favours  over , though nothing here hinges on that precise characterization of preference.
The Zelda case from above is then naturally modelled with cyclic preferences: Watson (1975, 210) in allowing these foreign forces to either be (i) forces that you do not identify with at all, or (ii) forces that occur with a strength or intensity that you do not identify with.This means that the addict who identifies with a desire for some drug, but not the overpowering cravings that drive them to relapse, counts as subject to a foreign force.These preferences suggest a simple explanation of my self-defeating behavior.On each day I compare the options available to me:  days of Zelda versus  + 1 days of Zelda and one less day writing.And on each day, I prefer the extra day of Zelda.Following my preferences on each day therefore leads me to something I wish to avoid-an unwritten paper. 3ut the explanation of self-defeat just given does not appeal to micromanagement.Micromanagement here is a matter of excessively tinkering with one's plans-"the procrastinator's failure to exhibit adequate self-control is often due to persistently second-guessing implementation plans that she appropriately adopted" (Andreou 2024, 177).And so to understand it we need to understand what plans are and how they enter the picture.
Plans enter the picture because agents with cyclic preferences must coordinate their choices over time.In the simple explanation of self-defeat in the Zelda case just given, I arguably adopt an objectionably narrow-sighted decision procedure.On each day I consult my preference over the options I face on that day, and I simply choose the option that I prefer.But since this day-by-day method of choosing dooms me to self-defeat, it is arguably irrational.Instead, rational agents with cyclic preferences should form plans that allow them to act in a dynamically unified way.Let us say then that I plan to play 10 days of Zelda and stop on day 11.
There are many theories of plans.Since the details matter in what follows, here is a representative sample: 4 3 Implicit here is that even if I have cyclic preferences  1 ≻  2 ≻ ⋯ ≻   ≻  1 , I can nonetheless judge that some items in the set { 1 , … ,   } are unacceptable to choose or end up with.For example, I might judge it impermissible to make extreme tradeoffs between work and play, which rules out the outcome '30 days Zelda (0 writing)'.While not every theory of choice will allow us to make this kind of claim (see Herlitz 2020), Andreou (2022, Chapter 3) develops an approach that allows for it.A complicating factor is the possibility of vagueness in determining which tradeoffs between work and play are permissible in the Zelda case.Allowing for that would require saying something about indeterminate permissibility, which would require more setup than I can give here.I therefore simply assume that in every set of options there is some subset of determinately permissible options and that you are required to choose or end up with one of them. 4I make no claim that this sample captures all the intricacies of various extant views in the literature.All that matters here is that they provide enough detail for us to understand how poor self-control interacts with planning.

RESOLUTE PLANNING CHOICE
The primary object of rational evaluation is plans.This means that an act is impermissible at some time if it is prohibited by a rationally-formed plan.Since I rationally planned to stop playing Zelda on day 11, then on day 11 I should stop playing.5 RESOLUTE PREFERENTIAL CHOICE Rational agents who make plans have the goal of steadfastness-they structure their preferences at each time around this desire to achieve dynamic unification (in particular, they disprefer options that are prohibited by rationally-formed plans).Since I rationally planned to stop playing Zelda on day 11, then on day 11 I prefer to stop playing and so should stop. 6LISTIC CHOICE Rational agents who make plans should adopt an evaluative stance that enables them to enact those plans.In particular, I am permitted to deviate from some plan for some reason at some time only if acting consistently on that kind of reason would nonetheless enable me to achieve an acceptable outcome.This means that in the Zelda case I might seemingly prefer playing an extra day on day 11.But I recognize that if this is a reason to deviate from my plan on day 11, my preference on day 12 is similarly a reason to deviate, and the same goes for each subsequent day.And so acting consistently on the kind of reason that speaks in favor of deviating on day 11 would doom me to self-defeat.I therefore evaluate options such that I stop playing on day 11. 7ch of these choice rules differs in important ways.But what matters here is what unites them.I will say that they all fall under: THE WEAK PLANNING PERSPECTIVE: Rational agents can make plans.If a rational agent makes a plan and later finds themselves deliberating in a situation envisaged by that plan, then absent a change in preferences or beliefs, they should carry out that plan.
The WEAK PLANNING PERSPECTIVE is quite weak, requiring constancy in only the best possible circumstances (the self who made a plan finding themselves in the situation they planned for).My own view is that it is hard to make sense of plans on anything less than the WEAK PLANNING PER-SPECTIVE: I am not sure how to interpret an agent who makes a plan but allows that they may rationally deviate from it in the circumstances most favorable to its implementation.
But even if you disagree with that, note that if the WEAK PLANNING PER-SPECTIVE is false, then we should reject cyclic preferences as irrational.
Various money pump arguments show that cyclic preferences can only be rationally maintained if agents can make and carry out plans. 8I therefore take the WEAK PLANNING PERSPECTIVE for granted here.
(There is a wrinkle-those who accept the WEAK PLANNING PERSPECTIVE can ignore it and skip to Section III.Andreou does not require that cyclic preferences are rational, so long as they are "familiar, understandable, challenging, and at least seemingly rationally innocent" (173n5).If so, then you might wonder whether we can really assume the WEAK PLANNING PERSPECTIVE on the grounds that rationality requires it.
I think, however, that in this context we must insist on rational cyclic preferences.This is because if such preferences are not rational, then one cannot properly identify with them.If I can see that my policy for trading between work and play in the Zelda case leads me to ruin, then I cannot endorse preferences formed on the basis of that policy as ones that I truly want in the driver's seat.In Frankfurt's (1971) terms, I desire that I desire differently.Or in Watson's (1975) terms, my values and desires do not line up.Either way, I am being motivated to act in the Zelda case by a part of my psychology that I do not identify with. 9And this means of course that we have not really escaped the foreign force paradigm.) 8See Gustafsson (2022, Section 4) for defense of this claim.See also Andreou (2023, Chapter 2). 9 Of course there is no single preference or desire that I fail to identify with here-it is my set of preferences in their entirety that I disavow.I do not think that this affects much.Consider an addict with a craving for three distinct drugs.Taking any two of those drugs is compatible with the addict living a balanced life, but taking three tips them over the edge.In this case there is no particular craving that the addict fails to identify with.Rather, they fail to identify with the set of cravings that in combination causes them to over-consume.And yet if this addict succumbs to temptation and takes three drugs, this seems like a paradigmatic instance of poor self-control to be explained by the foreign force paradigm.And so I think we should allow that one counts as subject to a foreign force even if there is no particular preference that one fails to identify with.It is enough to disidentify with a set of such preferences.In the Zelda case then, if I think my cyclic preferences irrational because they interfere with my ability to achieve important goals,

III. THE TENSION
At this point we have a tension (reminiscent of one that Andreou herself recognizes -see Andreou (2024, 1-172)).The micromanagement explanation of poor control is that, despite having made a rational plan, I engage in excessive deliberation such that I second-guess my plan and deviate from it.But we have just seen that, because of the WEAK PLANNING PERSPEC-TIVE, I must deliberate in the context of my overall plans.How then could excessive deliberation cause me to deviate from a plan?For example, how could deliberation be a bad thing for a resolute chooser who thinks about acts as parts of plans?On the face of it, the decision rules sketched above should ensure that deliberation over the course of a plan enables (or at least does not interfere with) that plan's completion.

IV. TEMPTATION
The resolution to this tension is obvious: temptation.When I procrastinate in the Zelda case, I am motivated on day 11 to do something other than I ought.
But this answer raises another question: what does this temptation consist of?
The natural thought in the Zelda case is that I in some sense recognize that I should stick with my plan on day 11.But plans are complicated and abstract kinds of things.And Zelda is playable on a shiny television, right here, and right now.The immediacy of the options in front of me renders them salient such that I am motivated to depart from the plan I ought to follow. 10ut this characterization of the temptation raises another question: when I am motivated to depart from a plan because of the immediacy of the options in front of me, am I under the influence of a foreign force?Or, recalling that talk of foreign forces is supposed to pick out parts of then I look at my cyclic preferences together and fail to identify with this overall pattern (or perhaps some feature of my psychology that is responsible for this overall pattern). 10Another possibility is that I am not thinking about plans at all on day 11.In other words, though I am a planning agent, I go into autopilot and 'forget' about the planning context I am in.This may very well be going on in some cases of procrastination, but it is not an adequate explanation here.For one, it would count not as a case of poor selfcontrol but of failing to attempt to control the self in the first place.And more importantly, it would not be a case of me deliberating too much-as the micromanagement explanation would have it-but too little.Indeed, the remedy for this kind of autopilot is to engage my deliberative capacities more fully at each stage of the plan.And this speaks against the kind of disengagement-based proposal Andreou (178-180) ends up endorsing.my psychology that I do not identify with: when I am motivated to depart from a plan because of the immediacy of the options in front of me, am I being motivated by a part of my psychology that I do not identify with?I answer yes.
You might be tempted to answer 'no' on the grounds that I depart from my plan because of my own preferences-the fact that I prefer 11 days of Zelda to 10 is by stipulation a part of a pattern of preferences that I fully endorse.And so I identify with the part of my psychology that motivates me to depart from my plan.
But on reflection we should reject this tempting thought.Agents who make plans evaluate decisions in the context of those plans.This means that motivational forces must similarly be considered by planning agents from the planning perspective.And from that perspective the agent cannot identify with a motivational force that causes them to depart from their plans.To make this thought precise, let us consider each of the decision rules sketched above.
First, say that I am a resolute planning chooser.Then on day 11 in the Zelda case I deliberate about the decision to play Zelda as part of my plan to stop on day 11.Do I identify with the part of my psychology-call it a desire, craving, inclination, or whatever you like-that motivates me to abandon that plan?Because a resolute planning chooser directs rational evaluation towards entire plans, they should think about their desires (or cravings, or inclinations, or what have you) in the same way.That is, when they reflect on a desire (or what have you), they ask how that desire functions in the context of a plan.If then that desire causes them to deviate from a good plan and implement a bad one, it thereby counts as bad and so something they desire not to have.If I deviate from my plan, I count myself as moved by a foreign force. 11econd, say that I am a resolute preferential chooser.My preferences on each day of the Zelda case reflect an overarching goal of dynamic consistency.Having this goal is consistent with my identifying with the desire to play Zelda on day 11.But it is inconsistent with my identifying with that desire to the extent that it overrides my goal of steadfastness.This is analogous to the addict who identifies with a desire to take a drug, but fails to identify with it to the extent that it overrides their ability to achieve other important goals in their life.And so again if I deviate from my plan on day 11, I count myself as moved by a foreign force.
Finally, say that I am a holistic chooser.On day 11 I do not take my desire to play Zelda as a reason for action, since counting that kind of desire as a reason for action would undermine my plan wholesale.But then, once again, to act on that desire would be to be motivated by something I disavow.And so, once again, we are back to the foreign force paradigm.
The common thread in all of this is that if I identify with my cyclic preferences, then I cannot identify with desires (or parts of my psychology more broadly) that undermine my ability to implement plans.We must therefore invoke the foreign force paradigm to explain poor selfcontrol-since I do not do what I want to do, it is no longer I myself who do it.

V. MICROMANAGEMENT REVISITED
Though I do not think we can do without the foreign force paradigm, I want to end by emphasizing that managerial skill nonetheless plays an important role in a constructive approach to poor self-control.
Plans help us achieve dynamically unified action.But of course we succumb to temptation, and we should make plans that anticipate those moments of temptation.12Cultivating a disposition not to micromanage our behavior, for example by developing appropriate habits as Andreou (178) suggests, then limits the opportunity we have to face temptation.
One problem with interpersonal micromanagement is that a manager's skills are not tailored to the specific tasks that their managees undertake.Managers should therefore stick to the higher-level tasks they are suited to.And I think that Andreou is right that the same goes in intrapersonal cases.When we deliberate we often find ourselves motivated by a range of (foreign) forces that, by our own lights, lead us to make poor choices.Our deliberative faculties therefore often operate best when we step back from the fog of individual decisions, which in turn often means evaluating courses of action before subjecting ourselves to the forces that may sway us while embarked on those courses of action.On the other hand, we are often surprisingly good at just getting on with a job we have committed to-writing a paper, for example-without deliberating much about it.And so by developing habits that limit micromanagement, we (i) make sure that our deliberative faculties are working on what they are suited for, while (ii) empowering our non-deliberative selves to get on with the job when needed.
Having said that, focusing just on managerial skill does not seem like it will lead to perfect self-control.Consider a long-term project like writing a book over a year.As before, while it is unacceptable for me not to write the book, I prefer on any given day to play Zelda over writing.I therefore make a plan to play Zelda once a week.Because this is a longterm plan, reasonable habits of reconsideration will allow that I reflect on that plan at least sometimes over the course of the year.But I therefore likely find myself reconsidering on days when I am inclined to procrastinate (and should not).And on these days I will need to empower myself to act on the desires that I identify with in the context of my plan.This is ultimately to say that the foreign force paradigm and the managerial paradigm dovetail.The former helps us understand what goes on in the moment of temptation, while the latter helps us to relate to and govern our selves who are tempted.Both strike me as important for understanding self-control.

VI. CONCLUSION
I have focused on two paradigms for understanding poor self-control, the foreign force paradigm and Andreou's novel managerial view.I have argued that the latter does not displace the former but complements it.And given that the managerial view is relatively new on the scene, I side with Andreou in seeing it as an especially fruitful tool for addressing problems of poor self-control-it helped me write this paper.
2Orthodox decision theory rules out cyclic preferences.It requires: Acyclicity: