Frosting the Glass (more on Euphemisms)

I ended the previous post (‘Public Labs’) with a quotation from the journal Comparative Medicine, chosen to show that ‘environmental enrichment’ – i.e. introducing natural or at least interesting features into the cages or boxes where laboratory animals are kept – has itself become a going animal-testing-equipmenttheme for research. But the quotation is also worth attention as a fascinating and instructive sample of animal-research prose. Here it is again:

However, animal wellbeing, as reflected by normal growth, development, and reproduction with low likelihood of injury, illness, distress, or maladaptive behavior, can exist even in housing situations in which the animal cannot perform its entire repertoire of species-appropriate behaviors, particularly if the animal will be maintained for a relatively short portion of its lifespan.

Let’s begin by appreciating that prize euphemism in the last clause, so rich in evasions that even now I don’t feel sure that I’m understanding it rightly. I take it to mean ‘particularly if you kill the animal unusually young’. If I’m right, this is something more sophisticated than the ordinary patch-words like ‘sacrifice’ or ‘euthanize’. A sort of smoke-screen has been laid down over the whole scene. The animal itself is, of course, made the victim of a passive verb (“will be maintained”): how often, in such journals, is a person ever seen doing anything to an animal in the active voice – assessing its ‘hot-plate latency’, for instance (see previous blog-post), by putting it on a hot plate? But, ingeniously, the verb in this case, though admittedly sharing in the general semantic fog, is detectably a beneficent rather than injurious one: ‘maintaining’ means looking after, doesn’t it? And it’s not even in the negative. In fact there’s no telling at what point in this clause the animal ceases to “be maintained”, a.k.a. is killed. The whole idea of time is helpfully obscured by converting it into space or quantity: ‘lifespan’ and ‘portion’. (For another instance of this same conversion technique, see the earlier post ‘Truths, Euphemisms, and Statistics’.) A hint seems to have been taken from the famous lines in the Victorian poet Arthur Clough’s ‘Latest Decalogue’:

Thou shalt not kill; but need’st not strive / Officiously to keep alive.

Clough’s poem is indeed largely concerned with what we allow others to see us doing. In that respect, the words of the article are the equivalent of well-frosted glass.

I don’t mean to pillory this text. It’s at least a more or less grammatical sentence, using unpretentious vocabulary. The trouble is that it’s so much in thrall to abstraction that the real and material subject – cages, and the animals inside them – is not so much illuminated as replaced by the words. Thus, instead of an animal being well, ‘animal wellbeing exists’; instead of a cage, a ‘housing situation’. In fact, instead of animals or an animal, that devitalized generic term “the animal”. And so on. The point is more simply made by translating the sentence back into real life:

However, an animal can remain in good health – that is, it can grow, mature, and reproduce, without injury, illness, distress, and neurotic behaviour – even in a cage where it cannot do all that it wants, particularly if it is not allowed to live long.

Has anything gone missing in this plain-spoken version? Perhaps the full sense of that phrase “low likelihood of injury …etc.”? But surely the sense is quite adequately expressed in the word ‘can’, meaning simply that it is possible, but not certain, that the animal will suffer none of these set-backs. The authors have presented this idea as a probability, implying that a known proportion of any collective of such animals will be free of illness. But in either version, the proposition only makes helpful sense if quantified, or linked to other studies which have quantified it, and this the authors do not do here. I would guess, therefore, that they have preferred “low likelihood” to “can” for the same reason that accounts for all the rest: it’s more abstract, further away from unscientifically particular animals suffering particular injuries – the equivalent in prose of cleaning up the cosmetic-testing-animal-remainsdisgusting mess on the work-bench before anyone else sees it.

The article in question is a review of other work rather than a report of original research, so there wasn’t any mess of its own to clear up; the sanitized style is really just professional habit. But it’s a thoroughly bad habit. No doubt we need euphemisms in our personal communications, for the sake of kindness and decency. And of course sciences all need their particular technical vocabularies, though probably not as much as their initiates like to suppose. But with the horrors of the 20th century laid out behind us, it hardly needs saying that – in public discourse – euphemism, and abstraction more generally, make life easier for every bad practice, from casual cruelty to mass slaughter. The more plainly we speak and write, the better we ourselves, as well as others, can see what we’re really doing, and whether we ought to be doing it at all.

I should briefly add that the article in question proposes that ‘enrichment’ is not the invariable good it might be supposed, for science or even for the animals. The study was connected, in some way not specified anywhere in the text, with the U.S. Army Research Institute of Environmental Medicine, at Natick, Massachusetts. That’s not an encouraging association (see the post ‘Remembering (some of) the Fallen’), but at least the article has been made freely available, and you can read it, if you wish, at


The photographs are by Brian Gunn (


Truths, Euphemisms, and Statistics

Thursday 22 October was publication day for the Home Office statistics of animal research conducted in the UK during 2014. BBC Radio 4’s Inside Science programme, broadcast on the same day, discussed the subject, and especially the inclusion in these statistics for the first time of information on the actual (as opposed to predicted) ‘severity’ of the experiments recorded. The presenter, Adam Rutherford, began by establishing the necessity of such research, its strict regulation in the UK, etc., so the programme’s point of view was made clear enough. Then he interviewed Dr Sarah Wells, Director of the MRC’s Mary Lyon Centre (mouse genetics) at Harwell. Dr Wells said that scientists themselves cat-animal-testing-pictureenthusiastically welcomed the innovation, and that the new statistics would be, for the general public, an “absolute true reflection” of the costs to animals of what happens in laboratories.

I’m sure she meant what she said about the enthusiasm, although as a matter of fact this kind of tonic response to public attention is what subscribers to the portentously named ‘Concordat on Openness on Animal Research’ promise each other to make (see their Annual Report, September 2015). But in the event neither participant could quite live up to the ideal of the absolutely true. When Adam Rutherford was asking Dr Wells to give Radio 4’s listeners an idea of what the different categories of pain implied, he seemed to feel that her truths had better fall some way short of absolute: “without being too graphic”, he warned her. And certainly her answer was reassuringly obscure. Her preferred word for pain of all types was ‘discomfort’ – plainly a euphemism when applied to anything worse than indigestion. When she came to define the ‘severe’ category, she blurred that somewhat unpleasantly evocative term by attaching it to the duration rather than the intensity of the animals’ suffering: “quite a severe period of time where they’re under discomfort”. It’s a strangely oblique, almost tortured bit of English, evidently the outcome of a struggle between candour and its opposite.

At the beginning of 2014, the Home Office published its own guidance on these categories specifically for the scientists. These Advice notes on actual severity reporting of regulated procedures are necessarily free of euphemism: free, that is, except in so far as the scientific outlook and terminology, having to be accurate at the expense of personal engagement, are themselves a variety of euphemism (“altered gait”, “autotomy”, “challenge with an inflammatory agent”, “repeated vocalisation”: yes, these surely are euphemisms, though with a motive behind them different from Dr Wells’s). Anyway, the Home Office text is surprisingly plain-spoken. Words like ‘pain’, ‘suffering’, ‘distress’ are used just as any reasonable person might use them of his or her own experience. In fact a reference to what we humans know of pain is indeed made at one point, when ‘severe’ is said to include “any state that a person would find difficult to tolerate”. It’s a great pity that this human reference is not used more in such discussions, but of course it violates a long and convenient tradition in science of resisting any suggestion that human experience can guide us in our understanding of animals. That would be called anthropomorphism, and accordingly unscientific. The way in which a quite proper scepticism has been stretched so as to justify denying to animals the rights of ‘painience’ (Richard Ryder’s term), makes an especially dishonourable theme in the story of animal research.

Still, a matter-of-fact bureaucratic survey, such as the Advisory notes provide, of all the varieties of suffering in laboratories (no, not all: suffering not caused by experiments, but by confinement itself, or by transport, or unintended illness, or fighting, or non-procedural accidents, etc., are not part of these returns) is liable to sound pretty heartless, and this one often enough does. See, for instance, a note on the ‘moderate’ category: “Pain of any significant intensity is of no more than a few hours duration.” Only a few hours? That’s all right then. Or “generalised seizures (in excess of one hour) with recovery will generally be considered severe.” There’s a history and prospectus of casual cruelty implied even in that one word “generally”.

You’ll notice that, in this last quotation, “recovery” seems to be regarded as compounding the severity, as well it might. And indeed failure to recover, a.k.a. death, is not regarded, in official animal-research ethics, as an existential evil, though it may be a professional nuisance: again, there’s a grim wisdom in that. The actual business of killing – the Home Office advice sensibly does call it ‘killing’, only once using the more refined ‘euthanasia’ – is expected to belongcosmetic-testing-equipment to the ‘mild’ category. (Let’s try not to picture those occasions when it strays into ‘moderate’ or ‘severe’.) But killing does not by itself count as a procedure at all. In fact the grand euphemism at work in every Home Office report on the animal research scene is the making invisible of this killing, not just of all or very nearly all the animals that feature in the ‘procedures’, but also of all the animals never used – the ones bred in excess of need, or found in some way unsuitable, and therefore dispatched uncounted. Yet much, perhaps all, of the mental distress felt by these animals while alive must consist in the very well-founded fear, however imperfectly understood, of premature death. Oxford University’s web-site boldly addresses this situation with its own prize-winning euphemism: “At the end of its life, the animal is humanely killed.” If only the animals themselves could read those consolatory words, and realize that they won’t, despite all their fears, be killed until the end of their lives!

Euphemism or heartlessness: it’s evidently a hard subject for practitioners to speak or write about without offending in one direction or the other. That’s a very strong indication, I would suggest, that there’s something wrong with the practice.