Hide table of contents

This article makes a case for selective contraception of wild animals as a near-term tool that could lead to a substantial improvement in wild animal welfare (hereafter, WAW). 

Based on my limited understanding of evolutionary biology, neurology and population dynamics, I argue that:

  1. Pleasure and pain are primarily a reward/punishment mechanism linked to evolutionary fitness.
  2. There is a neurological cost to producing pleasure/pain sensations. Highly valanced states (both positive and negative) can decrease evolutionary fitness.
  3. In a state of nature with a stable population, most species experience close to net zero utility.
  4. Without significant genetic engineering or ecosystem redesign, the best (only?) way to achieve significant positive utility is through ‘tricking’ evolution by using contraception; which will ultimately result in animals living longer.

I unpack these arguments below.

Pleasure and pain as a function of evolutionary fitness

Pleasure and pain are subjective experiences: I may feel pleasure at eating a brussels sprout, while someone else might find it painful. However, one strong theory explaining the origin and purpose of pleasure and pain is that they serve as a motivator for an organism to perform actions that increase evolutionary fitness and avoid those that decrease it. [1] Hence, most organisms seem to find activities like eating and mating pleasurable; while being too hot, cold, hungry or sick to be painful. 

The cost of pleasure and pain

Pleasure and pain serve a purpose, but they also have a cost. Producing hormonal signals takes biological resources. Probably more importantly, organisms that are experiencing extreme pleasure or extreme pain may well make worse decisions than they otherwise could. Consider the analogy of a rider (evolution) and a horse (individual organism). The rider can hit the horse with a stick (pain) to ‘push’ the horse forwards or change its direction, and can dangle a carrot (pleasure) in front of the horse to ‘pull’ it. But use of both the carrot and the stick costs some resources to the rider and distract the horse, so most of the time the rider lets the horse trot along, using pleasure and pain only sparingly. The costs of producing and experiencing pleasure and pain signals means that the default experience of a typical organism is close to zero.

Astute readers will notice my assumption: that the costs of producing & experiencing pleasure and pain are roughly symmetrical. If, on the other hand, it was much easier and less biologically resource intensive to produce pain rather than pleasure – then we would expect the ‘rider’ to use the stick far more than the carrot. The ‘default’ state for the organism would then be one of mild pain, and the ‘rider’ would reward it for good decisions that increase evolutionary fitness by stopping the pain. Conversely, if it is easier to produce pleasure than pain, we would expect the organism’s default state to be one of mild pleasure.

I have no empirical basis for this symmetry assumption. However, the case for contraception as a tool to improve WAW is even stronger if we believe that pain is easier to produce than pleasure; as contraception also reduces the overall population size. From a hedonistic utilitarian perspective, selective contraception would only be futile or harmful if it turned out that pleasure was much easier to produce than pain, and hence animals live on average joyful lives such that it is better to grow the population of animals substantially. This seems to me quite unlikely.

Most species experience close to net zero utility

Consider an organism’s life as a graph, with the x-axis being their age, and the y-axis being their evolutionary fitness score at that point in time. For a sexually reproducing species with a stable population, each organism produces on average 2 offspring. Hence at birth, our ‘mean’ organism has an evolutionary fitness of 2. It makes a series of decisions, and experiences different factors which increase or decrease its fitness. The derivative of this graph is roughly the organism’s current state of experience. 

Organisms who produce more than 2 offspring are likely to have experienced a positive net utility in life: they have lived longer, eaten and mated more. Organisms who die before producing offspring are likely to have experienced a negative net utility. This means that the whole species has a net utility close to zero. 

If this model is roughly accurate, it is an update in the positive direction: many EAs concerned about WAW assume that the natural world has a significant negative net utility. This is understandable when we consider that, for many species, the majority of individuals die a young and painful death, before having reproduced. The modal and the median organism may well have a negative net utility, but the mean organism is close to zero because of a minority who do live long and successful lives.[2] 

Contraception: tilting the scales towards pleasure

One of the main arguments deployed against concern for WAW is that it seems there is little we can do to avert animal suffering. With our current level of scientific understanding, most people are understandably reticent to attempt major genetic re-engineering, for instance, to make lions vegetarian. 

Consider short-term, technologically feasible and politically acceptable options. We can treat or provide food to individual animals, to help improve their lives. We can kill predators to temporarily improve the lives of prey animals (as controversially suggested by Will MacAskill). But ultimately equilibrium effects will bring things back to square one. By helping one wildebeest with a broken leg to survive, I am taking food resources away from other hungry wildebeest. By killing one lion, I may save tens of individual wildebeest, but do not change the carrying capacity of the ecosystem – an equivalent number of wildebeest will die from other causes. 

Even if we change the ecosystem substantially – for instance, by irrigating what was once a desert, to increase the wildebeest’s food source – there will only be a short period of positive net utility, while the population grows. Once the population stabilizes again, it is back to net zero utility, and the larger number of animals is meaningless. Malthus was wrong about humans, but he was right about animals. The repugnant conclusion is a matter of speculation for humanity’s future, but it is a present reality for most animals: their populations are as large as possible, while their welfare has been driven to zero.

But what if we gave animals the capacity to stabilise and control their reproduction rate? Contraception potentially enables us to trick evolution’s pleasure-pain symmetry, driving up the net utility of some species even while their population remains stable. 

For instance, consider a species with a population of a thousand males and a thousand females, which can live up to 10 years, and each mating pair produces one offspring every year. In the state of nature with stable population, there are a thousand births each year, which must be offset by a thousand deaths. Each individual lives, on average, 2 years. As argued above, the net utility is close to zero.

Now suppose we sterilise half the females.[3] The birth rate drops to 500 per year initially, while the death rate remains similar. The population drops, which soon improves the survival rate, as food becomes more abundant. The number of deaths drops (while lagging slightly behind the falling birth rate). The population ultimately stabilizes at half its initial size. The life expectancy doubles, which means that the lifetime fertility rate rebounds to 2.[4] 

As animals live longer lives, their net welfare increases because they are conducting more activities (eating, mating) which increase their evolutionary fitness. An organism that lives 4 years in the state of nature would have produced, on average, 4 offspring. The organism is thus likely to experience a positive net utility. [5]

A human analogy: the demographic transition

If this argument for contraception as a way to increase WAW seems surprising, consider the lived historical experience of human societies. Developed countries have undergone a demographic transition from an equilibrium state with high birth and death rates, to one with low birth and death rates. The precise mechanisms driving the transition are debated, and may vary in different times and places, but the overall thrust is summarised here:

  • The transition is initiated by improving healthcare and/or nutrition leading to a drop in death rates, resulting in rapid population growth (Stage 2 of diagram below).
  • There is then a time lag, as people realise that they no longer need to have four children in order for two to survive; now two will suffice. Hence they use contraception to lower the birth rate (Stage 3).
  • The birth rate stabilises to match the lower death rate, and population growth levels off (Stage 4).

You may have noticed the difference in causal order between the demographic transition in human societies and my proposal for wild animal sterilisation. In the demographic transition, drops in death rates increases demand for contraception, and leads to a drop in birth rates. Because the death rates fall first, with birth rates lagging, the population increases and reaches a higher new equilibrium. On the other hand, wild animal sterilisation leads to a drop in birth rates first, with death rates lagging: consequently, the population reaches a new equilibrium which is lower than the state of nature.

Comparison with Farmed Animal Interventions

Some interventions to improve the welfare of farmed animals seem staggeringly cost-effective, especially if we accept the headline figures of the Moral Weights Project: that welfare ranges of most farmed vertebrates are within an order of magnitude of humans, and the welfare ranges of invertebrates are within three orders of magnitude. While the Moral Weights Project focused specifically on animals that are farmed at large scale (or may be farmed in future), it is reasonable to assume that many wild animals which are biologically similar to their farmed counterparts have similar welfare ranges.

It’s true that work on farmed animals has a big advantage over that on wild animals: because we more directly control the lives of farmed animals, it is easier to change their living conditions and their manner of death. However, the third category of intervention mentioned in this paper, population size, is perhaps an area that is easier to work on wild animals. Interventions to reduce the population of farmed animals – such as a meat tax – would likely run into strong political opposition from both meat producers and consumers. In contrast, because there are no vested economic interests benefitting from wild animal suffering, interventions such as selective sterilisation are less likely to face strong political opposition. 


I acknowledge a major weakness in this article: I haven’t considered crucial ecosystem-wide effects. For instance, if we selectively sterilise Species A, would they simply be out-competed into extinction by Species B, which occupies a similar ecological niche but we have not yet sterilised? Or what would the flow-on welfare effects be for Species C, which preys upon Species A? This would clearly need careful analysis before conducting any such intervention.

This article was written from the perspective of hedonistic utilitarianism. There are other philosophical perspectives in which animal contraception could be viewed as harmful, for instance:

  • Eco-centric worldviews that emphasize the beauty of ecosystems in a ‘state of nature’ without human meddling.
  • Deontological worldviews which emphasize ‘do-no-harm’ and are thus averse to risky, uncertain interventions.
  • Rights-based worldviews that emphasize the right of individuals to not be forcibly sterilized without their consent. (how could an animal give this?)


I’ve argued that:

  • Pleasure and pain are largely a reward/punishment mechanism for inducing organisms towards behaviours that increase their evolutionary fitness.
  • Because of the cost of producing and experiencing valanced states, the default instantaneous experience for most individuals is neutral. In a state of nature with stable population, many species have close to net zero welfare.
  • We could potentially increase welfare sustainably and systematically through sterilisation, which results in lower death rates and ‘tricks evolution’ to break the pleasure-pain symmetry.
  • A similar phenomenon has occurred in most human societies, which have transitioned from a high to low birth and death rates, and enjoyed higher life expectancy and an improved quality of life.

Contraception is often mentioned as one of an array of measures to improve WAW, along with, for instance, vaccinations, protection from extreme weather events, and improving urban habitats for animals. However, I contend that, without contraception, these other well-intentioned actions are unlikely to make a significant net impact; as saving individual animals from one kind of painful death only increases population pressure and ensures a similar number of others die similarly painful deaths. Further, I believe that contraception can make a significant improvement in WAW even without these other well-intentioned actions, through the equilibrium effects described above.

I acknowledge that I have only surface-level knowledge of evolutionary biology and neurology, and that the models I’ve discussed here are highly simplistic. So I look forward to hearing your thoughts and critiques!

[1] Here evolutionary fitness means the number of offspring an organism is expected to have over their lifetime. More precisely, evolutionary fitness is about the number of copies of that organism’s genes in circulation into the long-term future. In other words, if Organism A produces 2 offspring which both die before reaching reproductive age, but Organism B produces 1 offspring that thrives and produces many descendants, then B is fitter than A.

[2] Another way to think about this is that since we are concerned about utility over time, the individuals who live longer ‘count for more’. For instance, consider a simplified species in which 90% of organisms die as one-year-olds, and experience an average instantaneous utility of -1 unit throughout their short lives. Meanwhile 10% of organisms die as nine-year-olds, and experience an average instantaneous utility of +1 unit. The species still has zero net utility.

[3] In reality, we would do this gradually to prevent significant population fluctuation. We would also need to continually sterilize a similar proportion each year.

[4] In my over-simplified model, the death rate is proportional to population squared (reflecting the fact that competition for resources intensifies as population increases). In this model, if we sterilize a proportion, S, of the initial population P, then a new equilibrium is reached with population (1-S)*P. The life expectancy increases by a factor of 1/(1-S). The proportions would vary if we used more complex and realistic models, but the overall direction of the effect – sterilization leading to decreased population size and increased life expectancy – is robust.

[5] The half who are not sterilized do indeed have 4 offspring on average; the half who are sterilized live otherwise identical lives but have 0 offspring. The net utility of both groups are positive; evolution has been ‘tricked’ into rewarding the population as if it is perpetually growing.





More posts like this

Sorted by Click to highlight new comments since:

Executive summary: Selective contraception of wild animals could substantially improve wild animal welfare in the near-term by reducing population sizes and increasing lifespans, thereby breaking evolution's pleasure-pain symmetry.

Key points:

  1. Pleasure and pain serve as evolutionary reward/punishment mechanisms, but producing them has a neurological cost. In a stable population, most species experience close to net zero utility.
  2. Without major genetic engineering or ecosystem redesign, contraception may be the best way to achieve positive utility for wild animals by "tricking" evolution.
  3. Sterilizing a portion of a population would decrease the birth rate, leading to reduced competition and increased lifespans. Longer-lived individuals would experience net positive utility.
  4. This is analogous to the demographic transition in human societies from high birth and death rates to low birth and death rates via contraception.
  5. Compared to interventions for farmed animals, contraception for wild animals may face less political opposition since there are no vested economic interests.
  6. Ecosystem-wide effects need further analysis, and non-utilitarian philosophical perspectives may object to animal contraception on various grounds.



This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.


I think this makes good sense as a toy theoretical model, and updates me some way towards these conclusions, but not very far because this sort of armchair theorising (while valuable and fun) is hard to get accurate for something as messy and empirical as this, as you note. So I think if someone were to investigate this further the key steps would be to:

  • look at empirical literature, or conduct primary research, on pleasure/pain symmetry and whether this holds (maybe this would be intractable though)
  • do some more involved population dynamics modelling, e.g. with a system of ODEs for food, prey juveniles, prey adults, predator juveniles and predator adults (I think this would be very tractable, but less crux-y)

That said, I think contraception as an intervention mode stands on its own without these more speculative theoretical arguments (unless wild animals have quite positive lives on average, such that preventing them from coming into existence is bad, but this seems unlikely).

Curated and popular this week
Relevant opportunities