On Noggle’s “Mistake Account”

Shlomo Cohen, Ben-Gurion University of the Negev

ORCID 0000-0002-9852-5221

Assigned DOI: doi.org/10.3998/jpe.9954

What is manipulation and what is its moral status? Robert Noggle’s Manipulation: Its Nature, Mechanisms, and Moral Status (2025) and my The Concept and Ethics of Manipulation (2025) both expound theories that answer these large questions. From a bird’s eye view, our answers are very different. Notably, Noggle provides a neat definition of manipulation, while I deny that manipulation is definable, seeing it as a cluster concept grounded in metaphor. In addition, Noggle views manipulation as pro tanto wrong, while I claim it’s not. This divergence in conceptualizations should not, however, mask extensive agreements regarding the complex phenomenon of manipulation. Importantly, it is precisely these agreements that prevent us from arguing past each other and that make the conversation fruitful. Noggle wrote an erudite, penetrating, and instructive book; here, naturally enough, I will focus on certain disagreements.

I.

Before addressing specific criticisms, a general point is in order. When different theories cover an extensive domain—such as the domain of manipulation—and when each is sufficiently comprehensive and internally coherent, it is far from clear that there can be a valid answer to the question of which is “the true” theory. Rather, the conceptual space may be sensibly carved in more than one way.

Inherent obstacles to providing “the true” theory of manipulation begin with the fact that different competent speakers use the concept somewhat differently—there is surely a legitimate range of linguistic sensibilities. In this vein we indeed find Noggle writing: “Some people define ‘manipulation’ more broadly, to include influences that do not seem pro tanto immoral. If you define ‘manipulation’ this way, please treat my use of the term as applying only to those cases of manipulation that do seem pro tanto immoral.”^[1]

Even if all had precisely the same linguistic sensibility regarding “manipulation,” however, other impediments to converging on the true theory would remain. The reason is that coming up with a theory of a concept is like attempting to solve one equation with multiple variables. There are multiple desiderata for a theory of a concept; these include: Loyalty to common usage of the term, a convincing phenomenological analysis, optimal internal coherence, a high level of parsimony, conformity with pre-theoretical moral intuitions (when applicable), and so on. Since we cannot realistically optimize all such parameters simultaneously, there can be multiple plausible theories of a concept. This is not, by the way, a peculiar limitation of “soft” concepts in the humanistic domains, such as the concept of manipulation. It is true generally and applies also to concepts in the natural sciences (e.g. “mass,” “species,” “simultaneity”), whose precise meaning (and reference) is a function of general theories about the relevant domains.

That the precise meaning of a concept is a function of multiple variables and of larger theories involving other concepts does not entail crude relativism, such that all theories of concepts are on a par. Each theory should attempt to convince why it is more compelling than the alternatives. Noggle provided an appealing theory of manipulation; I attempted one too. Whether they are on a par or not—time perhaps will tell. My task here is to point out some difficulties in Noggle’s theory, as I see them. Specifically, I will first present doubts regarding the identification of manipulation as such with causing mistakes, then I will criticize Noggle’s position about the normativity of manipulations/mistakes.

II.

According to Noggle, manipulations are successful instances of “acting manipulatively.” He defines the latter thus: “M acts manipulatively toward T if and only if M tries to influence T’s behavior or mental state by causing T to have what is, from M’s perspective, a mistake in T’s psychological states or processes.” Hence, in Noggle’s view, all manipulations produce mistakes in their victims intentionally. I disagree.

The concept of manipulation, I argue in my book, is constructed via the conceptual metaphor of “mechanification”: when certain parameters converge, influence starts being perceived through the image of the influencer operating—and thus controlling—his target as one operates a mechanism. The paradigmatic image of mechanification is arguably that of a puppeteer controlling his puppet. The puppet is moved so as to react or perform a role that the puppeteer wants it to perform. By harnessing the agency of his target, the manipulator-as-puppeteer makes the manipulee a pawn in his strategic plan. Now if the puppet-puppeteer (or more generally, mechanification) metaphor is what is constitutive of manipulation, then mistakes aren’t necessary. “The degree to which we perceive the target as being made to play a role in the influencer’s grand plan matters much for the identification of manipulation. [...] The more we see the target as ‘trapped’ in the role assigned to him by the manipulator in the game she is playing, the more likely we’ll see influence as manipulative.” (Cohen 2025: 108) Importantly for us, this need not involve any subversion of norms of rationality on the part of the target; “mistakes” are not essential.

When an influencer plans and executes a covert scheme in which another, in the role of her pawn, is maneuvered into a position from which she would choose in accordance with the influencer’s scheme, that is often enough to diagnose manipulation. The following example (adapted from Cohen 2025: 22) is of that sort:

Salami. Daphne wants to influence Gabby to choose D (a choice that would greatly benefit Daphne) instead of A (a choice that would not benefit Daphne). Gabby is indifferent between A and D, and so there is a 50-50 chance that Gabby would choose either option. Daphne is anxious about the possibility of Gabby choosing A, and is committed to do what she can to have her choose D. Weighing her options, Daphne concludes that the “salami technique” is her best chance to get Gabby to choose as she wants her to choose. Daphne therefore plans and executes the following scheme. She explains to Gabby the advantages of choosing B; this sounds reasonable enough to Gabby, and she indeed chooses B. As Daphne has anticipated, once Gabby has chosen B, it gradually starts seeming sensible to her to advance to C, and with a little rational persuasion from Daphne, she indeed does. Finally, after Daphne allows Gabby enough time to get used to C, moving from C to D, again does not seem an unnatural move. Daphne is of course there, ready to provide Gabby with relevant arguments in favor of D; convinced, Gabby chooses D. Daphne rubs her hands in satisfaction: her grand plan regarding Gabby, of which Gabby remained oblivious all along, is now accomplished.

In Salami, Daphne makes Gabby choose as Daphne wants by devising a covert strategy that maneuvers Gabby into a position conducive to choosing option D, through steps that are each and all rational. Gabby’s reasoning involves no mistakes. (See also the example called “Shrewd Beggar,” ibid.)

Not only is it not necessary for manipulation that the target makes a mistake, but neither must the manipulator intend to cause (what is from the manipulator’s perspective) a mistake in the target’s psychological states or processes. Bullshitting (in the sense popularized by Harry Frankfurt [2005]) is a case in point. The bullshitter attempts to influence his target to do or believe something by whatever means the bullshitter finds useful. He harbors no specific intention to cause a mistake in his victim and sometimes indeed he doesn’t. And yet, the insincere scheme and the sheer fact of caring only about the causal effect on the other make many cases of bullshitting qualify as manipulations.

According to Noggle, there are two kinds of manipulations, involving two kinds of mistakes: Manipulations can either be forms of trickery, where the mistake is typically obvious, or they can be forms of pressure, in which case the mistake is that involved in akrasia (weakness of will). Noggle explains that whereas cases of coercion present us with compelling reasons to submit to the coercer’s threat, pressure-manipulation presents no such reasons. The victim’s succumbing under the pressure is thus explained by akrasia. Noggle’s akrasia argument is insightful and deserves more examination than space allows. Here I will merely argue that succumbing to manipulative pressure need not be the result of akrasia. The following example is from Anne Barnhill (2014: 54):

Your partner wants to go on a family camping trip, but you don’t. While you’re discussing it, your partner calls out to your children, “Hey kids! Who wants to go on a camping trip?” The children cheer. You correctly judge that it’s better to go on the camping trip (despite its drawbacks) than to disappoint your children. You agree to go on the camping trip.

The partner is manipulated into agreeing through (social) pressure. And yet the partner’s reaction is not akratic—it is the result of sensible, cold calculation. Pressure-manipulation through “architecture” of the environment need not involve the mistake of akrasia.^[2] In fact, some pressure manipulations can even prevent akrasia (see the example “Roommate” in Cohen 2025: 142-143).

III.

Noggle’s view that manipulations induce mistakes and that they are therefore pro tanto wrong owes to his seeing a deep parallel between manipulation and deception. Noggle saw this parallelism as pivotal to the understanding of manipulation already in his first, important paper on manipulation (Noggle 1996), and now he has buttressed it with an instructive insight. Noggle writes: “if a case of manipulation can be redescribed as a case of deception, and if it's implausible to think that this redescription changes its moral status, then we can use our moral intuitions about deception to guide our assessment of the original case of manipulation.” This is an interesting argument, and it has prima facie plausibility. But while I agree that there is substantial affinity between deception and manipulation, and that we can learn from the former about the latter, I also think that the difference between them is significant enough to warrant caution about according the similarity a pivotal explanatory role.

If all manipulations were deceptions, then we could make straightforward inferences from deceptions to manipulations. However, such is not the case, nor does Noggle claim it is. Noggle rather claims that all deceptions are manipulations, and it is much less clear what precisely can be inferred from this about manipulations generally. Admittedly, if all deceptions are manipulations, then deceptions may be viewed as paradigm manipulations (for they are invariably manipulative), and as such they must be instructive. However: (1) This is compatible with the possibility that there are other paradigms, from which we may infer alternative lessons. (2) I don’t in fact think that all deceptions are manipulations (I cannot here elaborate but see Cohen 2023; 2025: 30-32; admittedly, though, deceptions are typically manipulative, and so this consideration is probably not decisive all by itself).

Noggle’s inference from deceptions to manipulations gains much force from his view that “to describe an (intentional) influence as inducing a false belief is to describe it as a case of deception.” I do not think this is true. There are manipulations that cause false beliefs intentionally yet that are non-deceptive. One can for instance present accurate information yet offer such an abundance of it that the listener will likely be confused and consequently form the false beliefs intended by the influencer; or one can present information fully and accurately yet frame it in a way conducive to making wrong inferences. In many such cases the delivery of the whole truth and nothing but the truth precludes deception, yet the delivery is done intentionally in a manner conducive to inferring falsehoods. These are cases of non-deceptive manipulation that cause false beliefs (Cohen 2018a). If all manipulations that cause false beliefs were indeed deceptions, this could support the alleged parallelism between deceptions and manipulations and a more plausible inference from the former to the latter. I think that is not the case, and so we should be cautious about inferring from deception to manipulation. More specifically and importantly, if some manipulations that cause false beliefs are non-deceptive, then it becomes much harder to be convinced that we should view manipulations as pro tanto wrong based on the analogy to deception.

IV.

Noggle formulates his basic thesis on the moral status of manipulation thus: “I contend that the normative status of manipulation rests on the normative status of mistakes. Manipulation is bad because mistakes are bad.” Suppose we accepted that manipulations invariably induce mistakes, what normative conclusions could we draw?

As mentioned, one of Noggle’s main theses is the parallel between manipulation and deception. Deceptions intentionally induce false beliefs; false beliefs are intrinsically bad mental states; inducing them intentionally is consequently intrinsically wrong. Noggle applies this argument to the inducement of mistakes by manipulation too. I have already mentioned that the parallel is suspect; now I want to focus on the very conception of normativity assumed by Noggle.

According to Noggle, “the fact that believing something false can be less bad overall than knowing the truth does not mean that it’s not a misfortune. It just means that sometimes it’s the lesser evil.” On “believing something false” Saint Augustine wrote: “when I put before my mind’s eye the intellectual beauty of Him from whose mouth nothing false proceeded, then, although my weakness reverberates in palpitation before the radiance of the truth shining ever more brightly, I am so inflamed by love of such great beauty that I despise all human considerations that call me back from there.” (Augustine 2014: section 36) Leaving theology aside, Augustine affirms the supremacy of the “intellectual beauty” of truth over any humanly consideration that might be thought to override it. Truth is beautiful and godly; the value of shunning error surpasses all mere earthly considerations. Unlike Augustine, Noggle admits that sometimes leading people into making mistakes is “the lesser evil,” and this is, I believe, a tremendous improvement. However, like Augustine (and many others since), Noggle too seems committed to the idea that inducing false beliefs is as such (pro tanto) morally wrong.

The foundational importance of truth notwithstanding, I must disagree. The fact that a false belief is as such epistemically bad does not entail that inducing one in another is (pro tanto) morally wrong. Hence, I do not find the “Augustinian picture” of the wrongness of deception convincing. Deception is indeed pro tanto morally wrong, but that is for reasons other than that error is essentially epistemically bad. Some hold that what makes deception wrong is that it undermines respect for the other’s autonomy, others because it uses people merely as means, yet others because it breaches a social contract, etc. I take it that deception is pro tanto morally wrong because it constitutes a breach of communicative trust. The existence and level of the wrongness of deception thus fluctuates with the relevant breach of trust (Cohen 2018b), not directly with the errors created. The distinction between deception and keeping in the dark is instructive (Carson 2010: 53-56). Both can create the same false beliefs and even intentionally so, but we decide to not call one of them “deception,” because “deception” is strongly associated with wrongness, and those behaviors that we classify under “keeping in the dark” often seem, and indeed are, morally unproblematic: there are many scenario-types in which withholding information does not breach trust, and is morally legitimate or, indeed, even praiseworthy. Hence, the epistemic norm—which should arguably issue the same verdict in both types of cases—is not here decisive. Rather, we have norms of moral interaction, and they determine our moral judgment, and simultaneously our analysis of moral concepts, independently of the question of whether we knowingly lead others astray epistemically. Moral normativity is (at least partially) independent from epistemic normativity. What may be epistemically criticizable may not be morally criticizable. (Exploring this further involves complex meta-normative issues that lie beyond our scope.)

Even if we did accept the inference from epistemic badness to moral wrongness for beliefs, its extension to other mental states and behaviors could be unwarranted. Noggle writes: “Something has gone wrong when an emotion is unfitting, excessive, or deficient for the circumstances. Something has gone wrong when someone fails to pay attention to something important. Something has gone wrong when someone acts against her judgement about what is best to do. Having something go wrong is a misfortune, even though there can be worse misfortunes.” While these are indeed norm-ruled behaviors, their connection to moral normativity specifically is yet more obscure. I do not see why inducing an unfitting emotion (say) should be intrinsically morally wrong. People welcome unfitting emotions when, for example, they watch a thriller anticipating they’ll get scared despite the safe environment of home or when they engage in role playing to see things from an exciting new perspective or perhaps simply for fun. Morality need not have anything to do with this. “Mistakes” do not seem to be as such intrinsically morally bad.

Although “mistake” clearly means falling short of some norm, mistakes are part of the good life. My discussion of how positive manipulations, which indeed frequently induce such mistakes, are integral to the good life (Cohen 2026) illustrates this point clearly. Furthermore, "mistakes are the portals to discovery," as the saying goes.^[3] Mistakes are basic learning tools. They help us refine our strategies and can induce more creative thinking. They have been the unavoidable path to many celebrated discoveries—from penicillin and radioactivity to LSD and corn flakes. No less importantly, they can build character through cultivating resilience and humility. Never making mistakes is clearly bad for us. Importantly, there is more than a consequentialist point here. It is misleading to view mistakes merely and invariably as a necessary evil, for mistakes are integral to the noble activity of learning. It is impossible, even in theory, to disentangle serious learning or training from mistake-making.^[4] Being curious creatures, experimenting is part of our identity; hence making mistakes is part of our identity, and it is counterintuitive to view this as intrinsically bad. Again, that mistakes are inherently bad under some norm (qua deviations from norms of truth, of prudence, of established practices, etc.) does not mean that inducing them is inherently bad morally speaking.

V.

Perhaps the real locus of my disagreement with Noggle lies somewhere else. Noggle writes: “But why are mistakes misfortunes? The answer lies in our status as rational agents.” I agree with Noggle that the ground for viewing mistakes as intrinsically bad (or not) relies on our conception of human being. Mistakes are often instrumentally bad for any creature, but for rational creatures they are also non-instrumentally bad. The crucial point here, in my mind, however, is that we are not merely rational creatures, we are more than that. Mistakes do not therefore automatically translate to misfortunes. I elaborate on this important point in Chapter Five of my book.

Noggle concedes that “some instances of manipulation are morally trivial.” He provides excellent examples too: “creating a false impression of enthusiasm when teaching dull material, or using gentle peer pressure to encourage a shy friend with a great singing voice to join in at karaoke.” I agree about the moral triviality of those cases, but then why insist on classifying them as anyway pro tanto wrong? Why not say simply that they are of no moral concern? Why argue that simulating enthusiasm to create a contagious effect in one’s students (which then typically rubs off on the teacher and creates a self-fulfilling prophecy) is a case whose “immorality is sufficiently minor to be easily outweighed” rather than say that there is nothing immoral whatsoever in such cases of skilful teaching? Indeed, is it not deeply counterintuitive to judge a practice as pedagogically laudable yet intrinsically morally wrong?

I find this position of Noggle’s confusing not just regarding this example but more generally. Noggle writes (quoted above) that his delineation of the notion of manipulation restricts it to those cases that do seem pro tanto morally wrong, yet he concedes that some manipulations are “morally trivial.” This strikes me as walking a tight rope, for how can the same cases seem at once both morally trivial and pro tanto wrong? Noggle provides relevant arguments for all his moves and, as we’ve seen, this apparently finally rests on philosophical anthropology—on a conception of human being as a rational creature. My own theory of the moral status of manipulation, too, rests heavily on a philosophical anthropology, albeit an alternative one. I find it interesting if the debate about the basic moral status of manipulation would be decided ultimately by the philosophical anthropology one favors.

In my book I wrote that the analysis of the moral status of manipulation might seem like a niche concern, but that it “taps directly into a profound predicament of the human condition” and so can be “an important window to the ethics of human relations” (2025: 119). In line with this diagnosis, we have seen in this short discussion how manipulation can be an intriguing test case not only for specific moral disagreements but also for our conception of moral normativity more generally, as well as for its grounding in philosophical anthropology. In such grand matters we can hardly hope for knockout arguments. My attempt here has been to indicate how things look from my perspective.

Bibliography

Augustine (2014) “Against Lying” [Contra Mendacium]. In Moral Treatises of Saint Augustine, translated by C. L. Cornish. New York: Aeterna Press.

Barnhill, Anne (2014) "What Is Manipulation?" in Coons, C. and Weber, M. (eds.), Manipulation: Theory and Practice, 51-72. New York: Oxford University Press.

Carson, Thomas (2010) Lying and Deception: Theory and Practice. Oxford: Oxford University press.

Chialvo, Dante and Per Bak (1999) "Learning from mistakes" Neuroscience 90.4: 1137-1148.

Cohen, Shlomo (2018a) "Manipulation and deception." Australasian Journal of Philosophy 96.3: 483-497.

Cohen, Shlomo (2018b) "The moral gradation of media of deception." Theoria 84.1: 60-82.

Cohen, Shlomo (2023) "Are all deceptions manipulative or all manipulations deceptive?" Journal of Ethics and Social Philosophy 25.2: 282-306.

Cohen, Shlomo (2025) The Concept and Ethics of Manipulation. Cambridge: Cambridge University Press.

Cohen, Shlomo (2026) “How to Think about the Basic Moral Status of Manipulation” Journal of Practical Ethics.

Frankfurt, Harry (2005) On Bullshit. Princeton: Princeton University Press.

Noggle, Robert (1996) "Manipulative actions: A conceptual and moral analysis." American Philosophical Quarterly 33.1 (1996): 43-55.

Noggle, Robert (2025) Manipulation: Its Nature, Mechanisms, and Moral Status. New York: Oxford University Press.

Noggle, Robert (2026) “The Mistake Account of Manipulation” Journal of Practical Ethics.

Steinberg, Elizabeth, et al. (2013) "A causal link between prediction errors, dopamine neurons and learning." Nature neuroscience 16.7: 966-973.