I must have struck a chord somewhere, for I have since been interviewed
by BBC radio about the article, and there have been one or two interesting
letters sent to me by readers.

Almost immediately, for example, a librarian by the name of A. Smith
in the Edward Boyle Library at the University of Leeds sent me details of
a mathematician by the name of Paul A. Smith (no relation) of Columbia University,
New York, who in the 1940s posed a difficult mathematical question which
was only answered in part in 1978. Such was the importance attached to the
question by the mathematical community that an entire symposium was devoted
to it in 1979, the proceedings being published in 1984 by Academic Press
under the title The Smith Conjecture. So now you know!

Meanwhile, Dr S. M. Lewis has written to me from Stockholm, and has
clearly been doing some research into the matter. Taking as his reference
guide the 1987-88 issue of the Nobel Foundation Directory of all prizewinners,
he first attempts to confound my hypothesis by pointing out that the directory
contains one Smith, two Lewises, two Jensens, two Simons, and four Wilsons.
Other examples that he quotes such as two Lippmans (different spelling),
Hauptmanns (ditto), Mistrals and Tinbergens I am not so sure about because
I don’t know how common they are in their countries of origin, but the point
about the others is well taken. To illustrate how common the name Lewis
is, Dr Lewis then says that there used to be 13 pages of Lewises in the
London telephone directory.

I do not dispute this fact, but it misses the central feature of Bagnall’s
hypothesis, which does not deny to the Smiths and Lewises of this world
the possibility of having laws named after them or winning prizes. Rather,
the hypothesis arises from the simple assumption that the number of Smith’s,
Jones’s, White’s, Lewis’s and so on laws should be in proportion to the
frequency of the surnames in the general population. Except that they don’t
appear to be, at least not to my knowledge, and I am puzzled as to the reason
why.

I am indebted to Dr Lewis, though, for pointing out some of the more
interesting names from the Nobel directory. Did you know that Quasimodo
won the 1959 prize for literature? Or that Millikan and Mulliken won the
chemistry prizes in 1923 and 1966 respectively? That Bothe and Bethe did
the same for physics in 1954 and 1967? That Muller and Muller did the same
for physiology/medicine in, respectively, 1946 and 1948? Or that Granit
(physiology/medicine 1967) was followed by Stone (Economic Sciences 1984)?
Pure coincidence or mystic force at work? Who can tell?

Michael Miller from Australia has also taken me to task, pointing out
such things as Miller indices, Brownian motion, the Richardson effect and
Young’s modulus. Again, this misses the central feature of Bagnall’s hypothesis,
that statistically these surnames should be doing better than they seem
to be. The most interesting name which Miller offers, however, is Murphy,
since it is quite possible in this case that the number of laws greatly
exceeds the statistically expected quantity!

But Miller’s correspondence explores another and possibly very fruitful
line of inquiry about surnames in science by pointing out that an unusual
surname doesn’t necessarily guarantee freedom from ambiguity. As an example
he suggests the con-fusion that would arise if a scientist named Greenhouse
had an effect named after him, and for an encore he muses on the misunderstandings
that could have arisen if the Van Allen belt had been discovered by someone
called Fan or Chastity.

Now, the fact is that situations like this just don’t occur in science,
and Miller has possibly stumbled on just one of the many ‘filter’ factors
that contribute to Bagnall’s hypo-thesis by excluding certain names from
scientific fame. Perhaps one of the great unanswered questions of our time
might be this: is an unusual but ‘real-word’ name like Chastity as doomed
to scientific obscurity as common surnames appear to be, or is there some
unknown cosmic force preventing people with the name Chastity from ever
working on topics such as radiation belts?

I am in fact already aware of this problem. It must be galling to people
like Random, Period, Law and even Murphy to know that no serious scientific
rule is likely ever to be named after them. No one called Null could ever
manage a hypothesis without creating a new field of chaos all of their own,
and think of the havoc that the Lasts of this world could unleash on science
if they got into the nomenclature. And did the well-known Fehling test,
which probably sounds very odd to nonscientists, slip through the net on
its spelling?

While you ponder all this, give thanks that the Krebs cycle wasn’t discovered
by Debye or someone called Peddle, and that Bunsen rather than Fawlty invented
the burner. What do the unions have to do with unionised chemicals, and
what is so periodic about periodic acid?

You see, names and the fame that goes with them are part of the great
cosmic soap opera of science, in which outright coincidence merges inexorably
into divine luck. And here’s a final coincidence to tickle the neurons.
Newton and Einstein both worked on gravity and mass. And the English translation
of ein Stein is ‘one stone’. What’s that again . . . new TON and one STONE?
Mere coincidence or a cosmic connection stretching across time and space?
Could it be that the next major step forward in understanding gravity will
again require someone with a name linked to mass? Who knows? But if your
name ends in POUND or GRAM you would do well to think seriously about physics
– fame could arrive at the speed of light (but probably no faster)!

Bob Bagnall is Senior Lecturer in Dental Materials Science in Edinburgh
University Dental School.

Now this in itself is quite impressive. I remember once trying to see
if I could detect a particularly malodorous substance by sniffing the exhaust
gases from a gas chromatograph, but repeated sniffs seemed to dull my sense
of smell and I failed miserably.

However, the television programme struck a chord for a different reason,
because it brought to mind two quite delightful if apocryphal stories of
animals in the laboratory, and the havoc that they can wreak on scientific
reasoning. I heard one of them during my first job as a postdoc, while carrying
out industrial research on potential new anaesthetics. It seems that, some
years earlier, my employer had been developing an intravenous anaesthetic,
and, after the usual progression through mice, rats, rabbits and so on,
its potential for veterinary use was checked in larger animals, with apparently
particular success in a horse. One day, however, during routine tests on
the beast, the operator had difficulty finding a vein and stood back for
a moment to rest, whereupon the horse promptly collapsed on cue and looked
for all the world to be fully asleep. As they say in the entertainment world,
never work with animals!

The second story is more interesting because it is more directly relevant
to the television programme. While I was doing my PhD, rumour had it that
a group of postdoctoral chemists had decided to go it alone rather than
enter industry, and had set themselves up in Cornwall making valuable intermediates
for the chemicals industry. It seemed that at least one experiment was particularly
hazardous and likely to explode without warning, and proper safety equipment
was thin on the ground. Fortunately, although none of the humans could detect
any impending danger, one of them had a dog which would whine and run away
if the reaction went critical. If it did this, everyone used to follow it
– and quick.

Now, I’m not suggesting that everyone rushes out and buys a dog. For
one thing, I don’t know the truth of the story, but there are so many tales
of dogs being sensitive to tiny environmental changes that it does at least
sound plausible. Animal rights aside, the idea of a laboratory dog being
equivalent to a miner’s canary seems quaint. It just needs a lot more feeding
and leaves larger droppings.

Why stop there though? Putting the smellprint idea into the laboratory,
the dog could also earn its keep as a general laboratory analyser. As any
chemist can tell you, many substances have characteristic smells that, once
sniffed, are never forgotten and can be sensed among a myriad of other potentially
confusing smells. No doubt the range of smells detectable by animals will
be much larger than that for humans; and the sensitivity of the dog’s nose
is so much greater than that of the human one that perhaps dogs could be
taught to recognise hundreds or even thousands of laboratory chemicals at
one sniff. Taking this idea a stage further, I seem to recall that the smell
of a substance is linked to its vibration frequencies in the infrared spectrum,
and, as any chemist knows, these are characteristic of different chemical
groupings within a molecule. Perhaps then dogs could be trained not only
to recognise known chemicals by smell, but to identify functional groups
in unknown substances. Even the human nose isn’t bad at this trick, and
experienced chemists can often make a reasonable guess as to the presence
of, say, an unfamiliar ester by smell alone. Imagine then how much better
a dog would be at the same game!

A dog in the laboratory would be a little inconvenient, however, particularly
if it decided to exercise some of its less attractive social habits. How
much more convenient it might be to use something simpler like moths or
other insects whose whole lives revolve around detecting one or two pheromone
molecules. Strapping them to the end of a gas chromatograph and keeping
them well fed might prove a little difficult, however.

I may not be the first to suggest this. Currently I’m working on the
theory that the high winds which caused such chaos across Britain earlier
this year were the result of trained laboratory moths beating their wings
in unison somewhere on the other side of the world and disturbing normal
airflow patterns. I’m thinking of calling it the moth effect.

My own need is for a different sort of detector. I can cope with reactions
getting out of control, and careful attention to safety rules should prevent
hazardous substances entering the environment, but what really stumps me
is when a vital piece of electronic hardware like a computer decides to
stop working. What I really need is an animal that can tell me when a computer
is about to go on the blink, so that I can line up an engineer (like plumbers
and policemen, they’re never around when you need them). If it could also
point to the offending chip, so much the better.

There could be rich pickings here for the pioneering inventor. With
the whole animal kingdom to choose from, it should just be a matter of gradual
elimination to find a beast with sufficient sensitivity to the minute fields
produced by modern electronics. Some animals would have to be discounted
right from the start, of course: a cobra in the laboratory would cause more
safety problems than it would solve, and a cow would make a bit of a mess
of the floor. What is needed is some small creature which requires little
feeding, can be kept in a corner, and doesn’t smell, make objectionable
noises or have dirty habits. A goldfish would be ideal, but, in the perverse
way that science works, the best animal detector will no doubt turn out
to be something awkward like an elephant or a whale. Still, think of the
publications!

Bob Bagnall is a senior lecturer in the dental school at the University
of Edinburgh.

Immediately, academics throw up their hands in horror because this approach
regards any teaching contribution as worthless, and time spent thinking
about difficult problems is time wasted because the flow of publications
is interrupted. The result for many academics is that difficult or long-term
research may be discarded in favour of the mundane routine stuff which generates
unsurprising results from standard, and therefore already acceptable, experimental
methods.

Putting aside the argument about teaching and long-term research, there
is a fundamental problem in using publications to assess worth, because
the actual method of assessment is itself open to question. Simply counting
the number of publications is obviously flawed, since it fails to distinguish
between the mundane and the significant. Nor can any sensible method be
developed for doing so, because the significance of an apparently routine
paper may not be recognised for many years. Further, a stream of mundane
work is no guide to worth of any one individual, because a moment of inspiration
is not confined to those who have already published significant work.

However, bureaucracy is not so easily deterred. One popular way of weighting
papers is to attach more significance to publications in certain journals
which are regarded as more prestigious. For example, there is at least one
British university where publications at one time actually scored ‘points’
on a scale which varied according to the journal in question, where conference
papers and so on also scored ‘points’, and where the progress of one department
could be compared with that of another in terms of the total number of points
for the year. Unfortunately, this system is also flawed, as it implies that
the same paper acquires more significance when sent to a more prestigious
journal, even though the wording and the work remain the same! But let us
return to the simple case of officialdom deciding to count the number of
publications. The intention presumably is that academics will set to, increase
their work rate, and out will pour a plethora of papers. What can happen
in reality is that academics may adopt one of a number of ploys to foil
the plan.

First, all long-term research can be dropped in favour of anything which
rapidly produces publications. Then all data which have previously been
considered too ‘ordinary’ for publication may be taken from the filing cabinet,
dusted down and sent to one of the less prestigious journals where hopefully
competition for publication is less. Secondly, and more importantly, all
future work which might have gone into a single publication can be fragmented
into several parts, each of which will be counted as a publication in its
own right. This, of course, is a master stroke, because in one bound it
increases measurable output for no extra effort. The only difficulty, and
one which has already received attention in the literature, is the question
of how small an article can be and still be accepted for publication, giving
rise to the concept of the ‘least publishable unit’.

Wait, I hear you cry. Officialdom isn’t stupid, and all the assessors
need to do is read each paper and they’ll soon realise what is going on.
But now for the hidden master stroke! These are expert papers, written by
supposedly the most intelligent members of the population, in esoteric language
understandable only to other experts, all of whom are themselves busy churning
out papers to prove their own worths. There is simply no time for all publications
by every academic ever to be read, let alone be assessed against all the
other papers. Academics find it hard enough simply keeping up with the constant
stream of current work without retrospective analysis of each and every
worker.

Officialdom therefore tries shortcuts, the simplest of which is to count
papers. Even then, however, academics can keep one step ahead by means of
multiple-author papers, whereby two or more academics who normally publish
alone agree to include each other on their future papers, thereby again
increasing apparent output for no extra effort. Another ‘basic’ concept
arises here, that of the maximum author loading that any one paper can stand.
An issue of New ÐÓ°ÉÔ­´´ some time ago suggested that this was subject-dependent,
with papers on, say, particle physics being able to stand more than a hundred
authors, and papers in less glamorous and hardware-dependent fields managing
perhaps only two or three.

One way that officialdom tries to counter the multi-author ploy is to
allow only first named authors to claim a paper as their own. This implies
that the first named author is the most important, whereas my own experience
is that this is often not the case. For example, academics may put the name
of a research student first to acknowledge the practical work done and to
encourage the student to work even harder – nothing gives a lift like seeing
your name in print! The matter is further complicated by the fact that,
to avoid any indication of extent of contribution, some journals list authors
in strict alphabetical order. (In my case this is extremely acceptable!).

The point is that, whatever technique it employs, officialdom cannot
win. It is not dealing with ordinary members of the public, but with a supposedly
highly intelligent elite whose whole being is dedicated to challenging rules
and solving problems. Every time officialdom tries a method of assessment,
the communal academic brain can, if it wishes, overcome the problem in a
single step and with minimal extra effort. Meanwhile, the research ouput
suffers as workers become bogged down in writing up trivia.

More recently, officialdom has tried a new ploy which seems aimed at
the least-publishable-unit fraternity: measuring the length of an article
in terms of the number of pages. But anyone familiar with the journals available
knows that some are printed on large pages of about A4 size, whereas others
can be as small as A5. The same article in different journals can therefore
occupy different numbers of pages, and a few extra diagrams thrown in will
soon bulk out a paper to an extra page or two, particularly in the smaller
journals. Page counting is flawed in other ways too. For example, an article
of mine in these columns last year was of a size to occupy less than a page
physically, but by chance it started near the foot of one page and spilled
over to the next. As a referenced paper it therefore appears to be a two-page
publication.

Something is very wrong with a system which is so easily beaten. The
error may lie in trying to assess academics in the first place. Assessment
is part of the appointment system for academic posts, and it could be argued
that thereafter academics should be taken on trust to do their best for
their subject. With few exceptions, most of the academics of my acquaintance
are not freeloaders looking for an easy ride, but dedicated workers whose
main frustrations are the lack of research money, lack of opportunities
for senior positions as vacant posts are frozen or lost, and poor salaries
compared with those in industry.

What then will officialdom try next? It certainly won’t give up the
attempt to assess ‘worth’, and one obvious option is the Science Citations
Index, whereby the value of a paper is judged by the number of times it
is mentioned in subsequent papers in the literature. Well, anticipating
this I have a cunning plan. It requires a rather larger group of co-conspirators
than before, perhaps a dozen or so academics publishing individually in
the same subject area, all of whom agree to cite everyone else in all future
publications. Think of the number of citations after only a few generations!
Not only that, but the length of each paper will increase dramatically with
the need to cite everyone else, thus increasing the page count and hitting
officialdom from two sides at once!

It is possible to imagine other scenarios for assessment, together with
a suitable solution which outflanks the system yet again. As an academic,
I find the whole process difficult to accept. Ideas and concepts seem to
have no value, presumably since they are difficult to measure. The result
is that the emphasis in British academic institutions appears to be on quantity
without regard to quality. Unless this process is reversed Britain will
gradually lose what remains of its hard-won reputation in science.

Bob Bagnall is a senior lecturer in the dental school at the University
of Edinburgh.

In my idealised world as I imagined it as a wide-eyed undergraduate,
the principal function of academics was to puzzle over all those observations
which did not fit into our understanding of nature, to live life at the
very frontiers of knowledge where certainty gives way to doubt, and where
the purpose of experiments is to peel back the layers of uncertainty until
the underlying truth is revealed.

At heart, this is still my idealised view of science. Certainly, it
is a view shared by those members of the public with whom I have discussed
it. Academics are still seen from outside as highbrow intellectuals whose
working days are spent huddled in corners discussing the finer points of
current theories, interspersed with forays to the blackboard to write out
complex equations in an atmosphere of feverish activity, or to the laboratory
to carry out critical experiments to decide between opposing ideas.

Academic scientists are not seen as messing around with what is already
known, but rather worrying about and challenging the fundamental concepts
of our understanding of all natural phenomena. They are supposedly so consumed
by their subjects, so driven by natural curiosity, that the real world seems
somehow remote in the archetypal absent-minded-professor manner.

This is a view to which I partly subscribe. In my idealised world of
science, academic research is concerned with the explanation of unusual
observations in the natural world. The purpose of academic experiments,
be they so-called thought experiments or actual practical work, is to assist
the explanation or to provide a corollary to it. Established scientific
principles can then be used by industry to provide useful objects to make
life easier for all. In this way, academia and industry should work together,
yet remain essentially apart, so that academics can follow their natural
curiosity wherever it may lead. Academic science, with the exception of
academic medicine, is not about applied research (the true province of industry),
nor is it concerned with unnecessary exemplification of established principles
merely to produce publications.

In all my years in research, however, I have never once encountered
my idealised environment. I have never found groups of academics huddled
over anything, and blackboards often remain empty except when they are used
for messages to other members of staff and for reminders to turn off the
lights as you leave. Perhaps I am looking in the wrong place, but on occasions
I have managed to engineer my ideal world alone, and when I have done so
it is a magical thing, difficult to describe to anyone who has not experienced
it. When facts which moments before lay scattered like the pieces of a jigsaw
suddenly fall into place, it leaves you both stunned and elated at the same
time.

Academics that I have met over the years are largely indistinguishable
from their industrial counterparts, with about the same level of inquisitiveness.
From time to time, however, academia does produce major advances in basic
science of quite stunning brilliance, and to account for this I nurture
a secret suspicion that in those laboratories which are recognised as being
the leaders in basic research, circumstances have combined to bring together
a team of true academics, all driven by curiosity and a sheer unbridled
love of their subject, unfettered by thoughts of applied matters. However,
since I have never experienced such an environment, it may yet prove to
be just a figment of my idealised world.

You see, it follows from my idealised picture of the role of academic
research that all academics should be engaged in the pursuit of understanding.
On this basis it should be possible to divide any subject into those parts
which are understood (hopefully the majority), and those parts for which
current understanding is poor. The business of academic research should
revolve around the latter, although the former should not be immune from
occasional attack.

It follows from this that all academic institutes around the world should
be recognising the same set of ill-understood facts and should be striving
in competition to unravel these various mysteries. It further follows that
it should be possible to go to any science department in any university,
ask them to name the 10 most important research topics in their particular
field, and obtain the same list from all. I doubt that in reality there
would be much similarity between the lists, but if there is then a possibly
even more interesting question arises. How many of those listed by a particular
institute are being worked on at this moment, and if some of them are not
being pursued, then why not? The reality is different for a number of reasons.
For one thing, the prevailing view in Britain is that academic research
must be directed towards the needs of society, which in effect means that
much of it is seen as having value only if it is applied. In turn, this
means that research questions often begin with the word ‘How’, as in ‘How
can polymers be made biodegradable?’ or ‘How can car exhausts be made less
polluting?’ For myself, I would prefer to see them begin with ‘Why’, as
in ‘Why are barnacles able to adhere to rocks under water?’

The problem may be due partly to the fact that innate curiosity is not
a prime criterion in appointing academics, supposed ‘curiosity’ being measured
in the list of publications on a curriculum vitae. Largely, however, it
is due to another aspect of the prevailing academic climate in Britain,
that of the continual need of academics to prove their worth by publications
and grants. It is far easier to churn out applied publications because the
scope for topics is not constrained as it is with the search for hidden
truth. When, in addition, grants favour such applied topics, the climate
is surely right for a plethora of applied research which is better done
in industry, and for the wasting of academic talent.

The reputation of British science was not founded on applied research.
Rather, it was founded on the search for truth in the natural world, and
it is the names of the scientists involved which will live on long after
the applied scientists have been forgotten. Unless something is done now,
then in years to come people will look back on the 1980s as the time when
Britain squandered its scientific heritage in the name of profit.

It is not too late to change course, but it will require a fundamental
change of political will. For myself, I will continue to believe that the
true path of academic science lies in probing the limits of understanding.
To do otherwise would be a betrayal of our scientific past.

Bob Bagnall lectures in the Dental School at the University of Edinburgh.

Musing on this one day, it occurred to me how confusing it would be
if several scientists with a very common surname all had laws named after
them. Now, in Britain at least, Smith is one of the most common names, and
I began to imagine a plethora of Smith’s laws appearing in the textbooks.
How would we know which was which? Life could get tedious if Smith’s law
of thermodynamics could be confused with Smith’s law of conservation of
energy, or Smith’s law of genetics. The marking of examination answers beginning
‘Using Smith’s law, we can show. . .’ could be a nightmare. Thankfully,
no such surfeit of Smith’s laws exists, and we are indeed fortunate in having
such a diverse range of names at our disposal.

While this idea was still bouncing around in my head, a stunning fact
revealed itself to me. Not only was science spared a glut of Smith’s laws;
in fact I didn’t know a single one! Nor could I recall a Jones’s law, or
a White’s law or any other law associated with one of the very common surnames
in Britain.

This struck me as odd. Assuming the brains to be spread equally among
the population, the number of Smith’s laws should simply reflect the proportion
of Smiths. Pick up a telephone directory, however, and a different picture
emerges. Take, for example, my own telephone directory, for Fife, and consider
the case of Boyle. There are about 25 times as many Smiths as Boyles listed,
so perhaps there should be about 25 Smith’s laws to counterbalance the one
Boyle’s law. But there aren’t. Want another example? Try the name Newton.
According to my directory, there are more than 50 Smiths to every Newton,
yet it was Newton who made a name for himself. Similarly, there are 25 Smiths
for every Rutherford, more than 400 Smiths to every Dalton, about 13 Smiths
to every Maxwell; and Faradays and Darwins are so thin on the ground that
they don’t even have an entry.

Now, I haven’t yet fully formulated Bagnall’s hypothesis of scientific
eminence, but I am beginning to suspect that the likelihood of eminence
is inversely related to the frequency of surname. Whether or not there is
an exact mathematical form to the relationship, or whether it is merely
a trend, is unclear at this stage, and would require not only a thorough
investigation of scientists from Britain but also of those from other countries.

How common are the surnames of all the Nobel prizewinners, for example?
How many people have the name Mendeleev in Russia? How many Avogadros are
there in the world? (The answer is not 6.022 52 X 1023.) Is Dirac a common
surname? How many Turings are there? Is the world teeming with van der Waals,
and is the connection between them weak? The list is endless, but the results
may share a common thread.

I must admit that I am at a loss as to how to explain these observations.
However, I cannot imagine that the Smiths or the Jones’s of this world are
any less capable than anyone else. Perhaps, and I offer this only as the
merest suggestion, rarity of surname acts as a subliminal flag when a scientific
paper is read, marking it out as memorable for reasons other than mere excellence.
Then again, perhaps those with the rare surnames feel that they have to
try harder to stop themselves being swamped by the rest.

The relationship may also hold good for other fields of human endeavour.
Perhaps a close investigation of the worlds of politics or business would
reveal a similar trend in relating success with scarcity of surname, though
once again I offer the thought as mere speculation.

I could of course be mistaken in all this. However, if my chance observation
has any substance, then the advice to any Smith, Jones or White aspiring
to the dizzy heights of science is simple. Forget it! The odds are stacked
against you from the start. Your only hope of becoming a scientific superstar
is to change your name by deed poll to the rarest name you can think of.

And what about my own name? In my local telephone directory at least,
it’s even rarer than Newton. By my reckoning, therefore, Bagnall’s hypothesis
stands a reasonable chance of acceptance. Whoever hands out the fame, please
note that I’m sitting here waiting . . .

Bob Bagnall lectures in the Dental School at the University of Edinburgh.