Fishing for theories

A group of researchers has reached the conclusion that fish feel pain (1).

Scientists at Britain’s premier Roslin Institute and the University of Edinburgh have taken some trout, popped electrodes into their brains and made recordings while they poked the fish’s heads and applied acid and heat. They discovered that neurones fire and also observed that after acid or bee venom injection into the lips the trout rub their lips on to the gravel in their tank.

There are some novel and important findings described by Lynne Sneddon and her colleagues, but it is the above details that have been emphasised by both the press reports and the authors themselves (1).

As any child who has ever prodded a goldfish or banged on the glass at an aquarium knows, fish can and do respond to threatening activity, and it does not take much imagination to infer associated brain responses. The firing of neurones and alterations in behaviour, however, do not make for experience. A plant will bend towards light and single-celled organisms will dodge obstacles. Even a computer will start whining when its battery gets low and will shut itself down when it gets really low.

But nobody thinks that plants enjoy being in the light or that amoeba prefer not to bump their heads or that my computer is having a hissy fit because it is hungry. These things are the substance of experience, and experience is more than a response.

The scientists argue that the fish behaviour should be viewed as experiential because it was coordinated and directed towards the area of injury. This is certainly consistent with an experience of pain, but it is equally consistent with no experience. An aeon of evolutionary pressure has selected for ordered and regular behaviour that leads to positive outcomes, and this can look like purposive behaviour. For awareness to be inferred, however, there has to be more than a regular and coordinated response.

When I want to write a document I call up a word processing package and the computer delivers; it does not throw me a graphics package or another random program, and yet the computer has evidently no clue as to what it is doing. The computer does not need to understand what is going on to respond appropriately any more than the trout need to experience to respond defensively during an attack.

By concluding that the trout feel pain the authors are implying trout awareness but are not proving it. It is a sleight of hand that preys on the reader’s willingness to accept the conclusion as foregone. People feel sorry for the trout, certainly, but there is no reason to believe the trout feel sorry for themselves. Trout don’t have awareness. If they did they would have gotten together with piranhas and attacked fisherman en masse long ago. Fish don’t do this because fish are not aware.

The authors do concede that what an animal feels ‘is possibly nothing like the experience of humans’ but is nevertheless ‘no less important in terms of biology or ethics’ than human pain. This is a tremendous muddle, that may have significant consequences for the clarity and integrity of my research field.

The researchers are trying to have it both ways, arguing that fish pain might be less of an experience than human pain but that we should, regardless, treat it as being the same. Millions of fisherman and those who consume their goods might justifiably ask ‘why?’. If the experience of the fish is ‘nothing like the experience of humans’ then the authors should either provide an attempt at describing this fish experience and/or adopt terms that encapsulate the uncertainty.

There have been several attempts to describe the experience of animals, and often the authors simply give up, deciding that it is an impossible question to answer, or they reach the conclusion of no experience because of the lack of evidence demonstrating actual animal awareness (2). Those who cling to familiar terms to describe animal experience often fall back on some version of the muddle put forward by Sneddon and her colleagues. Mark Beckoff, for example, has written that ‘even though the experience of pain might not be the same across species, individuals of different species can still suffer their own type or version of pain’ (3).

I have no idea what this means. It could mean that animals respond in their own particular way, which is simply to say that animal species are different. Or it could mean that animals have some special mind stuff that they carry around with them. Perhaps what Beckoff and others mean is that while animals are incapable of understanding (or being broadly aware) they are capable of experiencing something like pain (or being narrowly aware). That is, whereas our awareness is a general phenomenon carried along with every experience we have, from hitting ourselves with a hammer to enjoying a sunny day, that of animals might be piecemeal including, say, pain, some other sensations, sights and sounds but not much else.

For each ‘piece’ to be comparable with our experience, however, they must be whole or available in some nearly fully formed state. If this is not the case we are logically left with at least diminished experience of less moral value, which Beckoff, Sneddon and others seem inclined not to believe, and these whole pieces must come inside a ‘mind’ that is not capable of our subjectivity.

But if these events, such as pain, are circumscribed and trapped inside the animal’s head how can the animal ever become aware of that particular experience, or any other? Humans use symbolic labelling to differentiate ‘pain’, ‘red’, and so on, and it is through this public process of externalising our thoughts and feelings that those thoughts and feelings come to have meaning to us (4). We become conscious. No animal has the kind of social network that we humans share, and no animal has ever displayed evidence of having the kind of communicative skills that awareness of pain or anything else implies.

We are left with a somewhat startling alternative. Either animals do not have experience and live their life in mental darkness, or they have experience available to them but choose not to use any symbolic form of expression. Such expression would seem, on the face of it, to have a tremendous utility for fish who might employ it to inform us how annoyed they are at having hooks hidden in their food.

The muddle that Sneddon and others engage in does not end with the strange understanding of experience; they also fudge the consequences. In her report, Sneddon points out that the researchers only used a small number of trout (six) for the bee venom part of their study because of ‘ethical reasons’. They presumably thought it was better to risk a falsely negative result than inflict bee venom on four more fish. It is unclear whether the six chosen fish appreciated this.

Regardless, the authors did think it was okay to remove, ‘via a suction tube connected to a vacuum pump’, large pieces of brain from 10 fish. I have no idea as to what preference a trout might adopt but, personally speaking, I would definitely prefer the bee venom to the vacuum cleaner.

The report’s authors also go on to state that it would be interesting to assess similar responses ‘in a tropical fish species’ and that future work ‘should examine the cognitive aspects of noxious stimulation to assess how important enduring a noxious potentially painful event is to the mental wellbeing of this species’. Prior to the current study it might have been suggested that such work was okay because fish do not feel pain but, by the authors’ own conclusions, surely such work cannot now be considered ethical. Either the authors have not thought of this or they do not believe their own conclusion – or both.

Behind the authors’ suggestion that trout have experience, however, lay some interesting facts. Sneddon and her colleagues demonstrated a response to noxious events that is absent in more evolutionary removed species of fish. Moreover, they demonstrated the existence of a particular type of pain fibre in the trout that is also evident in humans. The range of response in the trout, though, is greater than that in the human, and the authors provide an angle on the evolutionary differentiation of sensory systems. This is good stuff and should have been the cornerstone of their report, rather than all the fluff about pain.

By getting caught up in a muddle about what fish might experience the authors detracted from their own important findings as well as potentially rendering their future work open to ethical dispute. The fact is that we share certain physiological properties with fish that are useful to study and interesting as a reflection of our evolution. But as the philosopher Daniel Dennett has remarked, ‘our reasons aren’t the reasons of fish just because fish are our ancestors’ (5).

Stuart Derbyshire is an assistant professor in the University of Pittsburgh Department of Anaethesiology. He is a contributor to Animal Experimentation: Good or Bad?, Hodder Murray, 2002 (buy this book from Amazon (UK) or Amazon (USA)).

Read on:

spiked-issue: On animals

(1) See Sneddon LU, Braithwaite VA, Gentle MJ. ‘Do fish have nociceptors? Evidence for the evolution of a vertebrate sensory system.’ Proc R Soc Lond B, published on the Royal Society website;

Randerson J, Fish ‘capable of experiencing pain’, New Scientist, 30 April 2003;

Press Release: Trout trauma puts anglers on the hook, Royal Society

(2) Dennett DC. Kinds of Minds: Towards an Understanding of Consciousness, Basic Books, New York, 1996

Heyes CM. ‘Anecdotes, trapping and triangulating: do animals attribute mental states?. Animal Behavior 46, 1993: 177-188

Visalberghi E, Limongelli L. ‘Lack of Comprehension of Cause-Effect Relationships in Tool-using Capuchin Monkeys’ (Cebus Apella), Journal of Comparative Psychology 108, 1994: 15-22

Carruthers P. The Animals Issue: moral theory in practice. Cambridge: Cambridge University Press, 1992

Horgan J. The Undiscovered Mind: How the Human Brain Defies Replication, Medication, and Explanation. Free Press, 1999

(3) Beckoff M. ‘Animal reflections’. Nature 2002; 419: 255

(4) Malik K. Man Beast and Zombie. London: Weidenfeld and Nicholson, 2000

(5) Dennett DC. Darwin’s Dangerous Idea. Evolution and the Meanings of Life, Simon & Schuster, 1995