An objection that is sometimes raised against suffering-focused ethics is that our intuitions about the relative value of suffering and happiness are skewed toward the negative for evolutionary reasons, and hence we cannot trust our intuition that says that the reduction of suffering is more valuable and more morally important than the creation of happiness. My aim in this post is to reply to this objection.
Stating the “evolutionary asymmetry objection” in brief
The argument I will respond to goes roughly as follows: For biologically evolved creatures such as humans, the reproductive costs of losses (e.g. deadly injury) are typically greater than the reproductive gains of successes (e.g. sex). This asymmetry is plausibly reflected in our experiences such that we tend to feel and value suffering (as an intrinsic negative) more strongly than we feel and value pleasure (as an intrinsic positive). Yet we should not expect such an asymmetry to be found at the level of possible states of suffering and pleasure. Instead, we should expect the best possible pleasure and the worst possible suffering to be equally intense, and we should therefore expect there to be an axiological and moral symmetry between them. Or at least we should expect our better informed selves to endorse such an axiological and moral symmetry (e.g. if we were fully acquainted with the utmost extremes of pleasure and suffering).
Furthermore, the experiences of future beings need not be subject to the evolutionarily contingent asymmetries found in the experiences of biological beings, and hence we should be far more optimistic about the value of the best future pleasures, and about the total amount of future pleasure, than we are intuitively inclined to be.1
Questioning the first three premises of the “evolutionary asymmetry objection”
A straightforward way to reply to the objection outlined above is by questioning some of its individual premises.
Are gains and losses asymmetric in terms of human reproductive fitness?
First, one could question whether gains and losses are in fact asymmetric in evolutionary terms for humans. In particular, one could argue that the difference between “being average” and “winning big” (e.g. by gaining great power and having a disproportionate number of offspring) is larger than the difference between “being average” and “losing big” (e.g. being dead) in terms of human reproductive fitness.
However, in my view, this is not a wholly convincing argument, since having a disproportionate number of offspring would tend to require a continual investment (even if only in terms of sexual investment), whereas dying would be a one-time event that would always be highly costly. And this would arguably be even more true for someone who was very powerful and biologically prolific — the reproductive cost of a single “death event” would still tend to be much greater than the reproductive gain of a single “sexual event” (or a similar “success event”), as the former would preclude many instances of the latter.
So it seems that steering clear of the worst outcomes moment-to-moment likely was more important than was attaining the best, in terms of the impact that individual events had on the reproductive fitness of our ancestors, even if there are counterarguments that limit the expected magnitude of this asymmetry.
Are reproductive gains tracked by pleasures as “intrinsic positives”?
Another premise that one could question is whether the above-mentioned asymmetry between reproductive gains and losses is necessarily reflected or “tracked” by an experiential asymmetry in suffering versus pleasure. Specifically, is it plausible to claim that humans are motivated to avoid reproductive losses by pains as “intrinsic negatives” while we are motivated to achieve reproductive gains by pleasures as “intrinsic positives”?
An alternative model could be that we are motivated by various desires (or felt cravings, or needs, etc.) that animate both the avoidance of reproductive losses and the pursuit of reproductive gains (cf. drive reduction theory). On this model, there could still be an experiential asymmetry in the sense that the desire to avoid death or severe bodily harm may tend to feel stronger than does the desire to find, say, a high-fitness partner. (One can, of course, dispute whether that is actually true of human desires.) Yet such an experiential asymmetry need not imply any motivation to achieve intrinsically positive states, as opposed to a motivation to attain a more relieved or less needful state.
Should we expect the best possible pleasure and the worst possible suffering to be equally intense?
The desire-based model of motivation outlined above may also question whether the best and the worst possible states are equally intense. That is, if one holds that we are chiefly motivated to satisfy (more or less bothersome) needs and desires, rather than to attain positive pleasures, one may likewise argue that there are no pleasures “on the other side” of a perfectly content state. Similarly, some Epicurean views of pleasure hold that the complete absence of pain is the “limit of pleasure” (Sherman, 2017, p. 103). On such views, it makes little sense to say that the best possible pleasure and the worst possible suffering are “equally intense”.
In general, regardless of whether one endorses any of the views of motivation and pleasure outlined above, one can reasonably question whether the respective intensities of pleasure and suffering are commensurable, i.e. whether they are measurable by the same standard (cf. Knutsson, 2016).
Yet even if we grant that suffering and pleasure do have commensurable intensities, and if we further grant that the best possible pleasures and the worst possible suffering are equally intense, it still does not follow that there is an axiological or moral symmetry between them. I think we have strong reasons to reject such symmetries, as I will try to explain in the next section.
What justifies a moral symmetry?
My main reply to the “evolutionary asymmetry objection” is to ask what justifies the acceptance of any kind of moral symmetry or moral outweighing between happiness and suffering to begin with, even at the level of currently accessible states that are claimed to be “similarly intense”.
I would argue that there is nothing about experiential states of happiness (e.g. excitement, gratitude, amusement, etc.) that render them a truly positive counterpart to suffering, neither in phenomenological nor axiological terms (Vinding, 2022).2 Instead, it seems plausible that there are no experiential states above wholly undisturbed mental states (Sherman, 2017; Gloor, 2017; Knutsson, 2022). (For an elaborate defense of minimalist axiologies that reject any supposed symmetry between happiness and suffering, see Ajantaival, 2021/2022.)
In moral terms, there is the argument that the reduction of suffering is morally urgent, whereas there is arguably no moral urgency, let alone a similar moral urgency, in “correcting” a neutral absence of happiness (Vinding, 2020, sec. 1.4). Unlike the presence of suffering, the absence of happiness does not seem morally problematic, which means that failing to create happiness (that nobody needs) is akin to a victimless “crime”.
More generally, there are various arguments for the moral principle that it is wrong to create happiness at the price of suffering, and that happiness can never morally outweigh suffering (Vinding, 2020, ch. 3). These arguments include thought experiments in which the supposed moral symmetry between happiness and suffering would imply that it is morally right to torture some beings for the pleasure of others.
In my experience, proponents of a moral symmetry between happiness and suffering rarely address this implication, despite it being perhaps the most problematic implication of such a moral symmetry. And when the objection does get discussed, the main reply is often that one should not commit such torture in practice, or that the thought experiment is unrealistic (see e.g. Lazari-Radek & Singer, 2017, ch. 4). Yet this reply obscures the fact that a general moral symmetry indeed would entail this implication in theory, and even if we were to grant that the scenario is unrealistic, it still appears to be a highly implausible theoretical implication.
More than that, the reply obscures the fact that tradeoffs like these are realistic in terms of whether we prioritize preventing extreme suffering or whether we prioritize creating new happy beings, and thereby allow more extreme suffering to occur by omission. (Whether we ourselves impose or merely allow the occurrence of the extreme suffering in question does not ultimately matter according to strict consequentialist versions of the moral symmetry, e.g. the version defended in Lazari-Radek & Singer, 2017.) And note that a vast future indeed would contain large amounts of extreme suffering in expectation, even if we avoid the worst risks of astronomical future suffering, or s-risks (Vinding, 2020, sec. 14.3).
In sum, the objection outlined above seems to assume that suffering can be morally outweighed by pleasure, but it does not provide any justification for this premise. Yet that, to my mind, is the key premise that needs to be defended, especially by addressing its most problematic implications.3
Should we be humble given our narrow and potentially skewed range of experience?
A proponent of the “evolutionary asymmetry objection” might argue that our narrow and potentially skewed range of experience should make us humble and uncertain in our inferences regarding the moral (a)symmetry between happiness and suffering, and they might further argue that this uncertainty should push us toward the symmetric view. Yet this objection seems problematic for a couple of reasons.
First, it seems to overlook that we do have significant data to draw on from our current range of experience, and one may argue that we have good grounds for skepticism about the relevance of the “evolutionary asymmetry objection” based on this data. As one author put it (Anonymous, 2015):
Our current pleasure/pain-intensity ranges may be negatively skewed for evolutionary reasons, but this doesn’t provide a strong argument for people who are able to experience the most intense current pleasures and milder current pains and are convinced that there’s an asymmetry.
Anthony DiGiovanni makes a similar point regarding experiences of which he has first-hand knowledge (DiGiovanni, 2021b):
The problem is that when I try to compare apples to apples in terms of intensity of experience, I still don’t see how happiness (or complexity, knowledge, beauty, whatever) could compete with suffering for moral priority. I have the same intuition when I consider cases where the intensity of the purportedly positive experience is quite clearly higher.
I myself draw the same introspective conclusion (Vinding, 2022). And so does meditator Roger Thisdell, who reports having experienced many states of unusually intense bliss, e.g. “bliss trips, jhanas, 5-MeO, MDMA, staring into the eyes of a lover without insecurities, laughing fits”, yet he still argues that “pleasure as a positive … does not exist” (Gómez-Emilsson & Thisdell, 2021; Knutsson & Thisdell, 2023). Indeed, many traditions that have developed practices of careful introspection appear to have converged on similar asymmetric conclusions regarding the nature of pleasure and phenomenal value (see e.g. Contestabile, 2014; Breyer, 2015; Vinding, 2020, sec. 8.14).
Second, the objection above overlooks that the point regarding humility and uncertainty cuts both ways. That is, just as we are generally far from experiencing the greatest pleasure, we are likewise far from having experienced the worst suffering, and it is not clear whether most of us have been “closer” to the worst suffering than to the greatest pleasure. It is therefore questionable whether uncertainty in light of our narrow range of experience should ultimately push us further toward a symmetric view.
Indeed, if an Epicurean view of pleasure is correct or most plausible, our experiences will tend to be much further from the most intense suffering than from the most intense pleasure (i.e. the complete absence of all disturbances, Sherman, 2017, p. 103). The same would be true on the broadly similar views of pleasure defended by Arthur Schopenhauer, Eduard von Hartmann, and others. Thus, if one gives some weight to these views, and if one were to update one’s beliefs about the extremes of pleasure and suffering based on the various views that have been defended regarding the nature of pleasure and suffering (views that, crudely speaking, tend to range from “symmetric” to “strongly negatively asymmetric”), one could argue that we should overall expect our notional “distance” from the most intense suffering to be greater than our distance from the most intense pleasure.
After all, just as one might argue that we should not a priori privilege asymmetric views of the nature of pleasure and suffering, one might similarly argue that we should not privilege symmetric views either when reasoning about these matters from an uncertain perspective, least of all when many alternative views have been defended.4
Finally, the humility objection still does not address how the absence of any hypothetical state of bliss could be morally problematic or morally urgent to address if no existing being feels a need for it, let alone how its absence could be as morally problematic or urgent as the presence of a state of suffering. The non-problematic nature of the absence of a given state seems independent of the nature or intensity of that state (Vinding, 2020, sec. 1.4).
Counterbiases
The “evolutionary asymmetry objection” essentially claims that our assessments concerning the relative moral value of happiness and suffering are biased, and that we would endorse a moral symmetry if only we controlled for this bias, or at least we would get closer to endorsing a moral symmetry.5 Yet if we grant that such “evolutionary biasing” is possible, one could plausibly argue that we have various biases in the other direction as well, and that we are in fact overall strongly biased toward endorsing a moral symmetry — not necessarily in terms of assuming equal intensities, but in terms of assuming that pleasure can morally outweigh suffering in the first place.
First, as Thomas Metzinger has argued, it seems plausible that we have a strong “existence bias” that pushes us to favor existence at virtually any price (Metzinger, 2017):
I claim that our deepest cognitive bias is “existence bias”, which means that we will simply do almost anything to prolong our own existence. For us, sustaining one’s existence is the default goal in almost every case of uncertainty, even if it may violate rationality constraints, simply because it is a biological imperative that has been burned into our nervous systems over millennia.
This is related to Robert Daoust’s claim that humans tend to have strong survivalist intuitions and values that frequently override welfarist concerns (see also Vinding, 2020, sec. 7.11).
In the context of our appraisals of the supposed moral symmetry between happiness and suffering, one could argue that our existence bias and survivalist intuitions plausibly bias us toward endorsing a moral symmetry between happiness and suffering. That is, a moral symmetry conforms much better with our existence bias and our survivalist intuitions than does a moral asymmetry that favors the reduction of suffering, and hence these evolved intuitions plausibly push us strongly toward accepting the desired symmetry. And if Metzinger and Daoust are right about the strength of our existence bias, then this putative bias in favor of embracing symmetry may well be stronger than the purported “evolutionary asymmetry bias” against symmetry.
Another potentially biasing factor is that we are used to thinking in terms of positive and negative numbers in just about every sphere of life. Consequently, we might be inclined to reflexively assign positive and negative numbers to different experiences when trying to represent their value, even if this conceptual move may not be the most plausible way to represent value on reflection.
Likewise, one can speculate that we are inclined to project positive value onto those things that tend to reduce pain and frustration — e.g. things that have positive roles in the alleviation of suffering — while overlooking that this seemingly intrinsic positive value ultimately has its basis in the reduction of suffering and unmet needs. After all, it would be quite demanding if we were to unpack this positive value in terms of its relational roles, and hence the “intrinsic positive value” framing may be more practically efficient and adaptive, even if the notion of intrinsic positive value might not be axiologically plausible on a deeper analysis. (Arguments along these lines are pursued in Ajantaival, 2021/2022.)
All in all, if we grant that we can be biased in our reflections on values, it is far from clear that we are more biased to reject a moral symmetry than we are biased to endorse it, including — and perhaps especially — when we consider potential biases that relate to our evolutionary origin. Indeed, it seems that one could reasonably argue that we are overall more biased to endorse rather than to reject a moral symmetry between happiness and suffering. (For a review of other potential biases, see Vinding, 2020, ch. 7.)
Disproportionally intense suffering may persist in the future
Another relevant consideration is the empirical point that states of suffering might continue to be more intense than are states of pleasure (if we grant, for the sake of argument, that states of suffering and pleasure have commensurable intensities). After all, overridingly intense suffering appears to have been adaptive for biological beings in the past, which suggests that this pattern could also be adaptive for beings in future scenarios that involve similar processes of competition and evolution, even if those scenarios involve advanced technologies.
In other words, if the future will be highly competitive (as seems fairly likely), and if disproportional intensities of suffering confer adaptive advantages in competitive environments (as seems to have been the case historically), then future beings seem likely to also be motivated by disproportional intensities of suffering — perhaps even maximal intensities of suffering among beings designed to be maximally motivated.
To be clear, I am not saying that the future is guaranteed to resemble the past, but merely that the range of experiences that motivate sentient beings today does represent some evidence regarding the range of experiences that we should expect to be prevalent in sentient beings in the future. This consideration means that the “evolutionary asymmetry objection” gives us less reason to be optimistic about the future than one would be if one ignored the likelihood of competitive futures and the seemingly adaptive role of disproportionally intense suffering. (See also “Beware underestimating the probability of very bad outcomes”.)
Reply to excerpts from Shulman, 2012
The following are a couple of excerpts from Shulman’s “Are Pain and Pleasure Equally Energy-Efficient?”, along with my replies to them. Shulman’s essay seems to be the most cited exposition of (something in the ballpark of) the argument that I am critiquing here, and hence I find it worth replying to key parts of that essay.
Shulman first defines two quantities, namely “hedons per joule” of the state of matter that produces the most pleasure per unit of energy (which he calls H), and “dolors per joule” of the state of matter that produces the most pain per unit of energy (which he calls D). (As Simon Knutsson has stressed, it is important to be clear that these respective quantities are a measure of energy-efficiency and not intensity; after all, the pleasure intensity of the state of matter that produces the most pleasure per unit of energy could in principle be extremely low.)
Shulman then proceeds to write:
By symmetry, my default expectation would be that H=D.
In addition to questions about the nature of the “hedons” and “dolors” invoked in these respective quantities, it is natural to ask what justifies the assumption of symmetry (i.e. “by symmetry”). This key premise does not seem justified in Shulman’s essay. And given that there are many arguments against thinking of pleasure and suffering in symmetric terms (including at the level of phenomenology), it seems that at least a minimal defense is required (see e.g. Sherman, 2017; Vinding, 2022; Knutsson, 2022).
Shulman again:
In humans, the pleasure of orgasm may be less than the pain of deadly injury, since death is a much larger loss of reproductive success than a single sex act is a gain. But there is nothing problematic about the idea of much more intense pleasures, such that their combination with great pains would be satisfying on balance.
A key question in this context is what is meant by “satisfying on balance”. In particular, what does “satisfying on balance” mean when some beings, or individual consciousness-moments, declare their experiences to be so bad that nothing could ever compensate for them?
In other words, we should be clear that the criterion according to which any state of suffering can be offset by other states, such that the totality is “satisfying on balance”, is not a criterion that is in agreement with all the beings or consciousness-moments involved (Tomasik, 2015; Vinding, 2020, ch. 4; DiGiovanni, 2021a).
A criterion that admits of such a “satisfying balance” must forcefully override the preferences and value assessments of the worst-off beings and consciousness-moments who declare their experiences to be unbearable and unoutweighable by any purported positive goods. In contrast, abstaining from the creation of happiness (that is not needed or desired by existing beings) does not violate the preferences of anyone (DiGiovanni, 2021a).
Acknowledgments
For helpful comments, I thank Teo Ajantaival, Tobias Baumann, Jesse Clifton, Emery Cooper, Anthony DiGiovanni, Simon Knutsson, Winston Oswald-Drummond, and Brian Tomasik.
References
Ajantaival, T. (2021/2022). Minimalist Axiologies. Ungated
Anonymous. (2015). Negative Utilitarianism FAQ. Ungated
Breyer, D. (2015). The Cessation of Suffering and Buddhist Axiology. Journal of Buddhist Ethics, 22, pp. 533-560. Ungated
Contestabile, B. (2014). Negative Utilitarianism and Buddhist Intuition. Contemporary Buddhism, 15(2), pp. 298-311. Ungated
DiGiovanni, A. (2019). Trading for Happiness: In Defense of Suffering-Focused Value Theory. Ungated
DiGiovanni, A. (2021a). Tranquilism Respects Individual Desires. Ungated
DiGiovanni, A. (2021b). Empirical Asymmetries Do Not Explain Suffering-Focused Intuitions. Ungated
Gloor, L. (2017). Tranquilism. Ungated
Gómez-Emilsson, A. & Thisdell, R. (2021). A Conversation with Roger Thisdell about Classical Enlightenment and Valence Structuralism. Ungated
Knutsson, S. (2016). Measuring happiness and suffering. Ungated
Knutsson, S. (2017). Reply to Shulman’s “Are Pain and Pleasure Equally Energy-Efficient?”. Ungated
Knutsson, S. (2022). Undisturbedness as the hedonic ceiling. Ungated
Knutsson, S. & Thisdell, R. (2023). Roger Thisdell on undisturbedness, positive experiences, and the hedonic peak. Ungated
Lazari-Radek, K. & Singer, P. (2017). Utilitarianism: A Very Short Introduction. Oxford University Press.
MacAskill, W., Bykvist, K., & Ord, T. (2020). Moral Uncertainty. Oxford University Press. Ungated
Metzinger, T. (2017). Benevolent Artificial Anti-Natalism (BAAN). Ungated
Sherman, T. (2017). Epicureanism: An Ancient Guide to Modern Wellbeing. MPhil dissertation, University of Exeter. Ungated
Shulman, C. (2012). Are Pain and Pleasure Equally Energy-Efficient? Ungated
Tomasik, B. (2015). Are Happiness and Suffering Symmetric? Ungated
Vinding, M. (2020). Suffering-Focused Ethics: Defense and Implications. Ratio Ethica. Ungated
Vinding, M. (2022). A phenomenological argument against a positive counterpart to suffering. Ungated
- An argument along these lines is found in Shulman, 2012, though note that there are important differences between Shulman’s argument and the argument that I have outlined here. For example, Shulman is primarily talking about energy-efficiency rather than intensity or “best versus worst”, as highlighted in Knutsson, 2017.[↩]
- Note that one could think that suffering and pleasure have commensurable intensities without thinking that pleasure is a positive counterpart to suffering. One may think that pleasure and suffering can be comparably intense in orthogonal (rather than anti-directional) experiential dimensions, as it were.[↩]
- A common argument in defense of this premise is that most people are willing to accept tradeoffs between suffering and pleasure in everyday life. I have replied to that argument here. See also DiGiovanni, 2019; Vinding, 2020, sec. 2.4.[↩]
- Note that this point about uncertainty regarding the nature of pleasure is different from moral uncertainty. Indeed, one could argue that each of these forms of uncertainty independently push toward asymmetric views, given the range of views that have been defended at these respective levels. Specifically, in addition to arguments regarding asymmetries in the nature of pleasure and pain, there are independent moral arguments that further support giving greater priority to the reduction of suffering (Vinding, 2020, sec. 1.5, sec. 6.2; MacAskill et al., 2020, p. 185).[↩]
- For a discussion of what it might mean to be biased in our assessments of moral values, see Vinding, 2020, sec. 7.1. One could, for instance, define a moral bias as “a factor that influences our moral reflections in a way that we would not endorse if we were more fully informed”.[↩]