The topic of this review was suggested tome by the Council of the Inte
rnational Ethological Congress, who invited me, a non-ethologist, to a
ddress an international gathering of the most eminent of ethologists i
n Vienna, the birthplace of ethology, and call into question the relev
ance of science in general, and their scientific discipline in particu
lar, to one of their most ardently sought goals, namely the improvemen
t, through improved understanding, of the welfare of other sentient an
imals. My concern is to question the quality and utility of science in
general and ethology in particular as applied to animal welfare. This
topic has in the past provoked me to some severe criticism; for examp
le, 'A lot of well-intended welfare research is neither very good scie
nce nor very helpful to the animals.... Too much welfare research is t
in my opinion) flawed either because it is oversimplistic, or because
it is not so much designed to test preconceptions but to reinforce pre
judice' (Webster 1994. Dawkins (1997) has recently responded to this c
hallenge, addressing the question 'Why has there not been more progres
s in welfare research?' Her response is concerned largely with applied
ethology. My own criticism was not directed at ethologists in particu
lar. I was more concerned by the misuse of scientific method by those
who seek to obtain a so-called 'objective' measurement of something wh
ich they preconceive to be a stress (e.g. measurement of plasma concen
trations of cortisol or endorphins in animals following transportation
). Here the 'objective' measure frequently becomes the test that gives
the answer that they want, and if it fails, then they seek other 'obj
ective' markers until they achieve a set of measurements that supports
the subjective impression which they had at the outset. My second mai
n concern is that the welfare state of a sentient animal is a very com
plex affair and cannot be embraced by any single scientific discipline
, be it ethology, physiology, molecular or neurobiology. Unfortunately
it is also too complex to be embraced by a single-sentence definition
. The best I can do is to suggest that it is determined by the capacit
y of an animal to sustain physical fitness and avoid mental suffering.
The assessment of this is necessarily multidisciplinary.