a memoir by Martin Evans
“Publish or perish!”. The warning is more urgent in today’s world of science research than when it was first used in the nineteen-thirties. A young post-doc, hoping for a career as a research scientist, is expected to get about two peer-reviewed research papers published every year in a front-line journal. In addition he or she is expected to publish summaries, short reviews, or a chapter in a book. If one is a co-author within an active research team this is not too difficult, but is hard to achieve if one works more-or-less alone or has university teaching duties as well as a research project.
Anyone who regularly fails to meet these expectations will start getting comments, and perhaps verbal warnings from a superior. They will find it hard to raise the grants that fund most modern research. Their professor or head of department may sideline them for promotion and tenure. The head of their institute or college will begin to think about replacing them with someone more productive.
These facts will not be much in the thoughts of a young postgraduate, enjoying the excitement of laboratory research and library study for a doctoral degree. The satisfaction of having passed their first degree with a good grade, being offered a place to work for a higher degree, and the thrill of doing real research are enough. It is a special world that they have entered. They will still be under the guidance of a supervisor but are encouraged to think creatively of new ideas and test them out with scientific experiments. They may have to begin by acquiring special skills and familiarising themselves with the relevant publications of others, but after a while they will discover something new, something unique, a contribution to our total knowledge. However small it may be, there is no feeling quite so exciting and satisfying than finding something that nobody else has ever known. The never-ending joy of scientific research is that for every question to which an answer seems to have been found, a couple of further avenues for research will open up.
After a year or two, the young scientist’s life will be disturbed by their supervisor telling them that it is time for them to make their discoveries public. The first step will usually be a presentation of their work at a meeting of a scientific society, followed by publication in the society’s ‘Proceedings’. My own primary society was the Physiological Society, which holds several meeting a year in various British universities and research institutes. Those intending to present a spoken paper send in a short abstract ahead of time. It will usually be published afterwards in the Journal if the presentation is accepted at the meeting. This will often be a young graduate’s first publication in a science journal.
I recall reminiscing with some colleagues a few years ago, about the terrors of giving ones first paper. One of us had worked for her Ph.D. degree at a Scottish university. She still remembered her unworldly professor, with an unfortunate choice of words, telling her: “Miss S., I think it is time for you to expose yourself to the Physiological Society.” At any meeting of this Society one could expect to be addressing an audience that included several Fellows of the Royal Society, and perhaps one or two Nobel Laureates.
Many of us have had experience of meetings where the chairman had not enforced the time allowances, speakers overran excessively, and we were all very late for a meal break. Having been created originally as a dining society in 1876, when a few scientific papers were presented after the dinners, the Physiological Society enforces time limits rather rigidly. Ten minutes exactly are allowed for each speaker to present their paper, followed by up to five minutes of discussion time. Novice speakers, especially, are carefully rehearsed beforehand in front of colleagues to make sure that they can deliver their presentation entirely from memory, without reading any of it, within the ten minutes. I remember one meeting when Professor Davson, an eminent character in the society, stopped speaking in mid-sentence when the red light came on. The chairman had to beg him to finish, to the amusement of the audience.
In September 1968 I was a speaker at a joint meeting of the British and Italian pharmacological societies, held in Florence. The conference was held in the new Palazzo dei Congressi on the outskirts of that fascinating historic city, which still showed much evidence of the damage done when the river Arno had flooded the previous year. The new congress building had all modern conveniences: dimmable lighting in the auditorium, excellent projectors for the 35mm ‘slides’ that we then used to illustrate our research presentations, and simultaneous translation between the two languages. I stepped onto the dais during the first morning session, had been speaking for a couple of minutes and had just called for my first ‘slide’, when everything went dark and quiet. The entire electricity supply to that sector of Florence had crashed – as apparently it did from time to time. So I ad-libbed through my paper in total darkness, speaking loudly because the microphone was dead. I told them what I would have shown on the screen if only the projector had been working. I remember some sympathetic chuckles at this, coming from the darkness of the auditorium. I was well trained in delivering a talk from memory, since the Physiological and Pharmacological Societies absolutely forbid a paper to be read from a script. I received enthusiastic applause when I finished. I doubt that it was for the scientific value of the paper: it was as much for my punctuality in keeping to the ten minutes and for having coped calmly with the local problems. In due course my paper was published – without any mention of the power-outage.
Preparing a full paper for publication in a peer-reviewed journal has been transformed by the availability of computers and word-processing software, and internet connectivity. Nevertheless, authors still have to learn the conventions of writing for a scientific publication. The required style almost always demands a dispassionate tone, written in the third person. Authors are discouraged from phrases: “I set up the apparatus …” or “We noticed some unusual effects …”. Instead, one writes: “The apparatus was set up … effects were observed …” Emotive words must be avoided, and one learns not to use some common words that may have specific technical meanings in some contexts. For instance, ‘significant’ has a precise meaning in the statistical analysis of data. It usually implies a 95 percent probability that a finding is not attributable to random chance. It should be used carefully in general scientific discussions.
My first real job, as a young physiologist with a shiny new Ph.D. degree in the summer of 1956, was in a large department of a major research institute in London. My Ph.D. supervisor had heard of the new post through his network of friends, and most of the groundwork had already been done. My interview with the head of the department then took place in a corridor during a meeting of the Physiological Society. He asked me whether I knew what a ‘leaky condenser’ was. “Yes”, I answered (I had been trained as a radio maintenance specialist during my military service in the Royal Navy some years earlier). He then asked: “Do you get depressed?” “No”, I answered truthfully (though it was not true in later life). “Good. Can you start on the first of next month?” I was taken on to assist a talented medical doctor who wanted to research some applications of neurophysiology in clinical medicine. Angus was a wonderful character and, although my boss in formal terms, we became very close friends. He was something of a bohemian, working late, going home late, and dining between 11 pm and midnight. After dinner he and his wife would relax listening to eighteenth century classical music, jazz and Ella Fitzgerald, with generous glasses of cognac or burgundy. They would seldom go to bed before 2 am. Consequently he tended to be rather late for work. He drove rather fast in an old Aston-Martin and usually managed to arrive in time for the morning coffee break, thereby meeting the departmental rules. These were slightly unusual. When I joined the department I was told by the head’s secretary: “Doctor F. does not care what time you arrive for work in the morning, nor what time you choose to leave, as long as you are in the department for morning coffee.” I soon discovered the reason behind this: those of us who were able to leave our laboratories would gather for the coffee break in a large room. During this break some minor administration would be discussed and settled, avoiding the need for time-wasting formal departmental meetings. I commend this practice to all departmental heads!
Angus and I worked very well together. After some discussion we would jointly write up the results of our experiments. I cannot remember the details, but the writing would always be in longhand on sheaves of paper. We would discuss points of detail, and the manuscript would get over-written with deletions and insertions. We might then re-write it, again in longhand, before showing it to the head of the department. Dr. F. was an eminent scientist, who had come to Britain from Europe during the 1930s, and he was a rather autocratic head of department. He insisted on reading through all draft papers in detail, and would usually impose many changes. All authors were expected to wait in his office while he read and amended their drafts, and sometimes one would see a short queue of senior scientists in the corridor outside his office, waiting for their latest manuscripts to be edited and approved.
Much the same routine would take place in most academic and research departments, with the head checking the authors’ drafts before passing them to the departmental secretary or typist for the first (but probably not the last) typed version. The labour of typing, and often entirely re-typing, scientific papers was enormous. At that time few academics had any keyboard skills, and would depend almost entirely on a professional typist.
Most scientific papers present experimental results in quantitative terms, with the numerical data sets often supplemented by graphs or other diagrams. Even in the biological sciences numerical data were increasingly important in the post-war years, and applied mathematics was supplying methods of evaluating the numbers by means of statistical analysis. No longer was it sufficient to quote an average figure for some series of experiments: mathematical equations allowed the reliability of the average to be estimated, with calculation of ‘variance’ and ‘standard error’. If two sets of results were being compared, for two different experimental circumstances, journal editors were then demanding that the data sets were analysed with statisticians’ methods such as the “Chi-squared test” or “Student’s t-test” to prove that the differences were ‘significant’, i.e.: real and not due to chance variability. These mathematical techniques found particular application in the biological sciences, to help handle the natural variability which is normally much greater than in the physical sciences.
Research workers now have computer programs to do much more complicated statistical analysis of their data, and also to generate printer-ready graphs, histograms and other forms of illustration for a paper. In the 1950s and ’60s we would work away with slide rules and perhaps have access to a mechanical desktop adding machine. Later on more versatile mechanical calculators became available: I still have the beautiful precision hand-held mechanical calculator made by “Curta” which I had bought in the late 1960s. It was a silky-smooth product of the watchmaker’s art, held in one hand while the other turned a little handle on the top. It could simultaneously add up the sum of a set of numbers and the sum of their squares, for calculating variance. It was soon to be made redundant by the emergence of the early electronic calculators that could do as much, with less effort, though at first they cost about as much as the mechanically delightful “Curta”.
We would become skilled at drawing our own graphs in Indian ink on cartridge paper. Very few departments could offer a drawing or graphic arts unit to do this for us. We learned to use stencils with special pens to put neat lettering on our illustrations. Later on, lettering transfers in a variety of typefaces and font sizes became available and editorial boards began to specify what typeface should be used in the illustrations. Preparing a manuscript for publication thus included learning how to analyse our data with standard mathematical equations, and learning how to make neat illustrations.
When the typescript (usually in carbon-paper duplicate and sometimes in triplicate) and illustrations were ready we sent it all off to the editor of whatever scientific journal was deemed appropriate and waited. … And waited…
All reputable journals have arrangements for submitted manuscripts to be ‘peer-reviewed’ by independent experts in the appropriate subject. The names of these referees are normally unknown to the authors. Only recently has the balance been corrected by keeping the name(s) of the author(s) hidden from the referees. This two-way anonymity should prevent a referee from writing a hostile anonymous review of a paper by a rival with whom there might be professional disagreement.
Refereeing someone else’s manuscript is always a responsibility. A senior scientist in an active field might have a big backlog of other workers’ manuscripts waiting to be reviewed. It takes time to do the reading and cross-checking thoroughly, so it is often several months before an author hears from the editor. Most referees take the task very conscientiously, and it is not pleasant to have to recommend that an editor rejects a manuscript outright. I have done so once, and later wondered whether I was too harsh on someone whose paper was so far outside conventional scientific norms that I may have missed some fragment of gold among the rubbish.
Very occasionally one hears within two or three weeks that ones manuscript has been recommended without reservations. More usually it comes back with a few, or many, requests that it be changed for one reason or another. They might be minor changes in wording or layout, or the referee might ask for very substantial changes or even additions to the data or reworking of the data analysis. This is always depressing, and always involves unwelcome additional work. Occasionally the criticisms are unwarranted, coming from some referee who clearly knows little about the subject and has failed to grasp the point of the paper. In these cases it is usually possible to negotiate with the editor. In my experience, at least three-quarters of ones manuscripts are positively improved by the wise suggestions of unknown reviewers. But it almost always used to mean sending another amended manuscript back to the overworked typist!
Now most scientists will type their drafts and final manuscripts with some word-processing software and the burden on the departmental typist is much less. It also means that when the editor does require changes, these can sometimes be made relatively easily without having to wait for a place in the typing queue. More troublesome are the papers returned because a referee has decided that the data are insufficient to warrant the conclusions, and the editor then tells you that additional laboratory experiments are essential to meet the referee’s criticisms. This is always depressing news, and might mean the major interruption of a new sequence of research. It happened to me at the time I retired from physiological research. A referee had asked for additional experiments to supply extra data. By the time I received the report I no longer had access to any laboratory and it was impossible to meet the requirements. A big set of experiments were never published. Tough titty! I eventually forgot about it, and found entirely new areas of interest for my retirement.
Martin Evans was born in Wales but spent most of his childhood in South America, where his father worked, before returning to Britain with his mother aboard the SS Highland Princess in May 1945. He completed his two years of National Service in the Fleet Air Arm before entering Guy’s Hospital Medical School in 1949. During his training Martin decided to switch from medicine to the biomedical science of physiology. After obtaining a PhD in neurophysiology from London University he taught and carried out research in various research institutes and medical schools until 1987.
Martin Evans and his wife, Dr Janet West, who trained as a biochemist, share an interest in maritime history with a developing expertise in scrimshaw and maritime art. The couple maintains a list of the maritime museums of the British Isles, accessible at: http://people.ds.cam.ac.uk/mhe1000/