by Brian Tomasik
First written: 3 Feb. 2015; last update: 25 Apr. 2016
There's a reasonable possibility that insects have some degree of consciousness and can experience suffering. Given how many insects each of us harms or helps by our choices, consideration of insect suffering should play a significant role in our actions. For instance, we should generally avoid buying silk and shellac, reduce driving especially when roads are wet, and minimize walking on grass or in the woods. Most insect suffering results from natural causes such as predation, parasitism, physical injury, and dehydration. We should encourage concern for wild-insect suffering and research ways in which human environmental policies can reduce it. Our descendants should also think twice before spreading insects and insect-like creatures to new realms, which could multiply suffering manyfold.
Monks of the Jain religion are known for their compassion toward all creatures. They walk carefully to avoid stepping on insects and wear cloths over their mouths to prevent accidentally swallowing flies.
But can insects actually suffer from being crushed or inhaled? Or are these practices merely relics of a pre-scientific world view?
This piece reviews some evidence for why insects might indeed be conscious and feel pain, at least to some degree. It then considers the implications of taking insect welfare seriously, including the welfare of insects in nature. While most of us will not go as far as Jain monks, there are a few simple ways in which we can reduce insect suffering.
Insects can react aversively
The first question we need to answer is: How do insects respond to harmful or potentially harmful events?
As most of us have seen from direct experience, insects try to escape if touched or sometimes even approached. They struggle if trapped, turned over, or caught in a spider’s web.
Insects have numerous sensory systems, including for vision, smell, taste, touch, temperature, and humidity. While it’s sometimes claimed that insects lack pain sensors, these have been discovered in a few species of bugs, including fruit flies. And even insects that lack pain sensors specifically may still respond aversively to other kinds of stimuli.
Insects show negative reactions to, among other things,
As in humans, opiates can affect insect responses to pain. Crickets were slower to escape a heated box when given morphine, and this effect was blocked if the crickets were given the anti-opioid drug naloxone. The effect of morphine decreased over time (“drug tolerance”), and when morphine was stopped suddenly after four days of administration, the crickets jumped more aggressively in response to vibration than usual (“drug addiction”).
That said, insects do not necessarily show the same sorts of pain reactions as humans. For instance, insects don’t typically try to protect damaged body parts. They may continue normal activities in the midst of severe injury, such as when a locust continues feeding even while being eaten by a mantis. Different stimuli are painful to different organisms. As Jennifer A. Mather notes, “Because nuclear radiation can kill us without our feeling a thing, humans too do not always respond with pain to possible tissue destruction.”
Insects can learn
Fruit flies learn to avoid odors that have previously been paired with shocks and also seek odors that have previously predicted relief from shock. They can also learn to avoid one end of a chamber when going there is paired with heat and retain this memory after the heat is gone.
Bees can use odor to discriminate when they’ll be shocked for trying to feed on sucrose, avoiding or delaying extension of the tongue when shock is predicted. Bees also show “cognitive performances that were thought to occur only in some vertebrate species” including “sameness” and “difference” concepts. Bees can generalize rules from odors to colors.
Ants appear to remember the height constraints along the trail back to their nest. When a one-centimeter height limit was imposed on an ant path, ants cut smaller and rounder pieces of foraging material ahead of time in the foraging area, out of contact with the barrier that they would later encounter on the way home.
One study from 1986 concluded:
There is now no question, for example, that associative learning is a common capacity in several invertebrate species. In fact, the higher-order features of learning seen in some invertebrates (notably bees and Limax [slugs]) rivals that commonly observed in such star performers in the vertebrate laboratory as pigeons, rats, and rabbits.
In light of findings about insect cognition, Peter Carruthers argues that “many invertebrates [such as navigating insects] possess a belief-desire-planning psychology that is in basic respects similar to our own.” He adds:
Those who accept some form of utilitarian theoretical framework, in which the basic moral currency consists of frustrations and satisfactions of desires and preferences, will find it difficult to resist the conclusion that sympathy is owed to at least some invertebrates, just as it is owed to other human beings.
Cockroaches live together, recognize kin, and make collective feeding choices. When cockroaches are raised alone, they show “a behavioral syndrome induced by social isolation, similar to syndromes described in vertebrates” of less exploration, reduced feeding, less interaction if later put in contact with one another, and poorer ability to evaluate potential mates.
But are insects conscious?
Scientists distinguish the ability merely to sense and respond to pain from the ability to consciously feel pain. For instance, if you touch a hot stove, a reflex withdraws your hand before you’re consciously aware of what happened. Insect learning and cognitive performance help show that insect responses to harmful stimuli are not always just reflexes. But are insects conscious like humans and other vertebrates are?
Insects show simpler forms of several of the behaviors, abilities, and brain structures that vertebrates possess. So for many measures of what “consciousness” should mean, insects will tend to be weakly conscious according to those measures.
Several prominent scientific and philosophical theories of consciousness can be interpreted to imply that insects are at least slightly conscious. For instance, some neuroscientists hypothesize that integrating information is crucial to consciousness. All insects integrate information to some degree. Daniel Dennett’s “fame in the brain” model suggests that brain processes are conscious insofar as they have lasting impacts on later processes. This is true for some events in insect brains.
Bruno van Swinderen found selective attention in fruit flies, which he suggests might be one of the “remote roots of consciousness”. And these are the words of Christof Koch, currently Chief Scientific Officer at the Allen Institute for Brain Science:
We have literally no idea at what level of brain complexity consciousness stops. Most people say, “For heaven's sake, a bug isn't conscious.” But how do we know? We're not sure anymore. I don't kill bugs needlessly anymore. [...]
Probably what consciousness requires is a sufficiently complicated system with massive feedback. Insects have that.
Degrees of consciousness
If insects are conscious, it’s plausible to consider them “less conscious” than vertebrates. Humans have at least 100,000 times more neurons than most insects. But collectively, many insect minds may add up to something quite morally significant.
Moreover, insect brains pack more punch than their numbers of neurons suggest. Presumably intelligence has some correlation with consciousness, and sometimes intelligence is predicted based on brain size relative to body size rather than absolute brain size. One paper adds:
Neural network analyses show that cognitive features found in insects, such as numerosity, attention and categorisation-like processes, may require only very limited neuron numbers. Thus, brain size may have less of a relationship with behavioural repertoire and cognitive capacity than generally assumed [...].
From Charles Darwin’s The Descent of Man:
the wonderfully diversified instincts, mental powers, and affections of ants are generally known, yet their cerebral ganglia are not so large as the quarter of a small pin's head. Under this latter point of view, the brain of an ant is one of the most marvellous atoms of matter in the world, perhaps more marvellous than the brain of man.
In addition, if insects do have “experiences” of sorts, their subjective lives may proceed much faster than our high-level thoughts do because their smaller brains can complete a processing loop quicker. A trend of faster visual information processing with smaller body size has been found for vertebrates.
A precautionary approach
Some scientists are reluctant to attribute feelings to insects. They eye with suspicion the common tendency to anthropomorphize -- not just other animals but even trees, rocks, and teddy bears. Science proceeds conservatively, refraining from conclusions until a strong weight of evidence accumulates.
However, when making ethical decisions, we shouldn’t wait for conclusive proof. Rather, we should take precaution by assessing different possibilities and evaluating the consequences if each turned out to be true. Even if your probability for insects feeling conscious pain is low, if insects do suffer, the moral consequences are significant.
Entomologist Jeffrey Lockwood required his students to anesthetize insects before experimenting on them. He reasoned:
If we use anesthetic and it turns out that insects don't experience pain, the material cost of our mistake is very low [...]. However, if we don't use anesthetic and it turns out that the insects were in agony, then the moral cost of our mistake is quite high.
The American Veterinary Medical Association agrees:
While there is ongoing debate about invertebrates’ abilities to perceive pain or otherwise experience compromised welfare, the Guidelines assume that a conservative and humane approach to the care of any creature is warranted and expected by society. Consequently, euthanasia methods should be used that minimize the potential for pain or distress.
How humans harm insects
Some of the ways in which we hurt insects are obvious:
- Stepping on them, smushing them in garbage cans, leaving them trapped in the house to die on windowsills, and so on.
- Insects may drown in buckets of water.
- Walking on grass or hiking in forests may crush insects. (On average, a square foot of land contains 750 insects.)
- We may run over bugs while driving, especially during the rain when worms and slugs cross the road. (Worms and slugs are not taxonomically insects, but they’re similar in many ways.) Splatting against cars kills at least trillions of insects per year in the United States.
Less visibly, certain consumer items harm insects:
- Silkworms are boiled, steamed, or fumigated in the silk-production process, killing roughly 10,000 worms to produce one sari dress. Gandhi opposed conventional silk for this reason.
- The resin shellac requires 300,000 lac bugs for every kilogram produced. Most of the bugs may escape before shellac production begins, but some remain and get killed. Shellac is used in many products, including some furniture polishes, makeup, inks, and confectioner’s glaze on candies. (See here for which candies do and don’t contain shellac.)
Insects are also harmed as food:
- Many pet owners feed live insects to reptiles and amphibians.
- Worms are skewered as live fish bait.
- Humans eat insects in many countries, sometimes frying, boiling, or steaming them alive.
- Western hobbyist insect eaters may kill their food by suffocation, drowning, roasting, or eating raw. This video shows people mashing flies into patties and then frying them. The video claims that each patty contains 0.5 million flies.
Natural insect suffering
Even though humans kill at least trillions of insects per year, most insect suffering occurs in nature. The world population of insects is estimated at around 1 to 10 billion billion -- about a billion insects for every human. While some of these insects live for several years or even decades, most mature insects die within a few weeks or months (see p. 10 of this paper). According to Jeffrey Lockwood, many insects die from lack of food, water, warmth, and the like, while others are killed by “predators, parasites, and pathogens”.
In addition, most insects never reach maturity. Insects may have tens, hundreds, or more eggs per mother. In a stable population, at most two of these offspring can successfully reproduce on average. This strategy of “quantity over quality” for offspring is known as “r-selection”, and it implies that almost all insects die soon after birth. An example “life table” showing the mortality of corn earworm larvae can be found here.
Even if insect lives are on balance enjoyable prior to death -- a questionable assumption given the hardships of nature -- the experience of a possibly painful death after just a few days of life on average doesn’t strike me as a good trade. Of course, our assessment of the balance of pleasure vs. pain in insect lives is anthropomorphic. It may be that insects suffer relatively less than we would from comparable injury; one speculative reason this might be is that any given insect has a short future lifespan ahead of it and low odds of actually completing reproduction, so it should adopt a “now or never” risk tolerance and thus shouldn’t be held back by excess aversion to injury.
But there’s no a priori reason why life in the wild should be good on balance. As Richard Dawkins wrote in River out of Eden:
The total amount of suffering per year in the natural world is beyond all decent contemplation. [...] The universe we observe has precisely the properties we should expect if there is, at bottom, no design, no purpose, no evil and no good, nothing but blind, pitiless indifference.
Combining our best guesses of wild-insect welfare based on short lifespans with the seriousness and horror that extreme suffering entails, it seems safest to err on the side of assuming that insect lives in the wild are on balance painful and thus that it’s better if fewer insects are born, other things being equal. By analogy, would you start a pregnancy if it was very likely the resulting child would die soon after birth and would plausibly endure net suffering in the process?
Humans changing insect populations
A 2014 analysis tracked worldwide invertebrate (including insect) abundance and found that 67% of invertebrate types showed an average population decline of 45% in the previous 40 years. During the same period, the human population doubled.
Cities around the world cumulatively cover an area 1.5 times the size of India. All told, humans appropriate about 25% of potential plant biomass, which means less food to be consumed by insects and hence fewer insects born.
These trends of habitat and fauna loss harm present-day humans and future generations, through resource shortages and consequent social instability. However, the reduction in insect births that they imply is a silver lining -- one that prevents extraordinary amounts of short-term insect suffering. We should explore further whether there are other ways to reduce insect populations without so much of a human toll.
Unfortunately, not all human impacts on the environment reduce insect populations. Global warming is expected to increase overall insect abundance, because insect growth is closely correlated with temperature. That said, the net impact of climate change on all wild-organism suffering is less obvious.
I’m uncertain of the net impact that crop cultivation has on insect suffering. In the short run, crop planting (tilling and driving over the soil) and harvesting (cutting through insect-filled plants) crush lots of insects. On the other hand, cultivated land tends to produce less plant energy (and hence less insect food) per hectare than forest would, which may imply fewer insects born into short, painful lives.
The net impact of insecticides is also unclear. Insecticides kill most insects on a crop field, which might be thousands or millions of insects per hectare. At least in humans, insecticide poisoning can be relatively painful. On the other hand, all those insects would have died anyway in other possibly painful ways, many of them within a few weeks, and insecticides prevent them from bringing many more offspring into existence, most of which would have died a few days after birth.
In light of this uncertainty, it’s plausible that organic farming, which uses fewer insecticides, involves more total insect suffering, due to higher insect densities on crop fields. Of course, organic farms may use their own kinds of insecticides, such as Bt, which kills insects slowly over the course of several days.
Rather than eliminating insecticide use, we might aim to make insecticides more humane, by studying which kinds of chemicals or other agents kill more quickly or less agonizingly but no less effectively.
As a footnote, I should explain why I think insecticides might be net good but squishing healthy insects (e.g., on the sidewalk) seems plausibly net bad. The reason is that when a non-farm insect is killed, the food it would have eaten is left available for someone else to consume, so it’s not clear there has been any net reduction in insect populations by its death. In contrast, the food that crop insects would have eaten is fed to humans, cows, pigs, etc., which means that other insects won’t end up eating it (except some portion of the fraction that’s wasted). This reasoning relies on a model in which insect populations are proportional to food, which may not be accurate. We should develop a better understanding of insect population dynamics to inform policy choices like these. In the meantime, it seems like the most certain way to reduce insect populations is actually to reduce plant biomass, since this is the “root” of the food web.
The amount of potential insect suffering in the world is prodigious. Even if we only weigh insect consciousness in proportion to number of neurons, insects still dominate humans: Insects outnumber humans in population by about a billion to one, whereas human neurons per organism outnumber insect neurons per organism by only about a million to one. And as discussed above, it seems plausible that insects’ more efficient brains deserve greater ethical weight per neuron. In addition, insect lives may contain more extreme suffering per day than many human lives do, especially considering how quickly insects die.
We shouldn’t let the immensity of the problem lead to despair. As the boy says in a modified version of “The Star Thrower”, helping a single individual makes a huge difference to that individual.
In the short run, we as individuals can
- choose alternatives to silk (Note: I'm skeptical that "peace silk" is a good choice; I would incline toward non-silk fabric types)
- choose candies, cosmetics, wood finishes, primers, etc. that don’t contain shellac
- consider covering our lawns with gravel to reduce insect populations
- drive less, especially less when the road is wet
- seal our houses and garbage cans to prevent insect infestations
- talk and blog about the importance of reducing insect suffering.
In the medium term, researchers can
- study how much insects suffer due to various events, such as drying out, cold, predation, and insecticides
- examine how various human environmental impacts change net long-term insect populations (e.g., crop cultivation, cattle grazing, global warming)
- propose strategies for reducing insect suffering without too many adverse side effects on humans and other animals.
In the long run, humanity should
- incorporate insect suffering as an externality into economic and policy calculations
- engage in discussion about the relative importance of preventing insect suffering versus maintaining other values, such as biodiversity
- think twice before spreading wild-insect suffering to new realms, speculatively including other planets that may be terraformed in the coming tens of millennia
- consider the ethical implications of virtual insects, including insect brain “uploads” and insect-like artificial intelligences that will emerge in the next few decades.a
- How does the sentience of insects compare with that of present-day robots? I think the sophistication of almost all present-day robots is much less than that of almost all insects. One easy way to see this is that most robots can only do very narrow tasks, while a single insect can perform a wide range of behaviors: It can navigate environments, avoid injurious stimuli, find food, engage in mating, lay eggs, etc. Humans are far from building a robot that can do all of those tasks consistently in complex environments. Insects have been doing them for 480 million years straight. In addition, insects are at least billions of times more numerous than present-day robots. But in the long run, especially once digital intelligence colonizes other planets, the balance will plausibly shift toward robots being both more sentient and more numerous than insects. (back)