Hi! I’m back from the horror of book writing and have also finally navigated around the issue that the IT Help desk is 6000 miles and 8 time zones away from where I am!
Due to both my joyous enthusiasm at being back and Bernie’s increasingly bleak emails appearing like Marley’s Ghost and just saying “Blog!” I have decided to. So, (a bit like London buses, nothing happens then five come along!) you’ve got two blogs this week.
A couple of things have converged recently, firstly a conversation with some students and secondly, finishing reading an interesting, but rather sad book.
The conservation with the students centred around their third year dissertations, in other words their first go at real research in that they design, collect and analyse data without really being told what to do.
This is somewhat disconcerting for some students, as they’ve never ventured very far from a sort of preferred expository form of learning. In other words, they’ve been told how and what to do by their teachers. Being ‘out on their own’ tends to freak them a bit and leads to queues of students outside my office asking for ‘tutorials’ which becomes something of a euphemism for ‘help’ at this time of year.
The conversation I had with a few this week centred to their write-ups. We were talking about the presentation of data and I told them that their projects did not need to include any raw data, only those data that had been ‘processed’ in some way. In other words graphed, or statistically treated. I didn’t want pages and pages of numbers. I think I added something droll about having better things to do with my life and that I’d have more fun reading a telephone directory. This seemed to promote genuine surprise in the tutorial group. “What?!” they chorused, “No primary data?!!” That’s right I replied.
After a brief (stunned) silence, one of them then said “But, Rog, we might make it all up!” I was equally knocked back by this and after another stunned silence (on my part) I replied somewhat hesitantly, “But why on earth would you?”
We then entered into an interesting discussion about ‘making up’ data. I argued that to do so was to prove that you didn’t really ‘get’ science because that ‘finding out stuff’ was the best bit. They countered by implying that I was naive in the extreme and didn’t understand the pressure students were under. They may a point I suppose.
The second and related event was that I have just re-read an interesting book that came out a few years ago called ‘A Rum Affair: How Botany’s “Piltdown Man” was Unmasked’ by Karl Sabbagh. It’s a very good book and I recommend it, although be warned I am about to rather ‘plot-spoil’ it if you haven’t read it.
It concerns the work of two academics in the 1950’s in the UK. One was Professor John Heslop-Harrison, a professor in botany who worked at Newcastle University and the other John Raven, a fellow at King’s College, Cambridge, who was a classical scholar. Raven worked primarily on pre-socratic philosophy but was also a keen amateur botanist.
Heslop-Harrison had (essentially) at this time proposed the idea that certain plants may have survived successive ice ages in so-called ecological refugia. In other words places such as sheltered, south facing locations where micro-climates may have established the required conditions for their survival. In other words, plants normally associated with distributions that were much further south, could well be found at more northerly latitudes. It was and still is a popular idea but at the time needed evidence to really verify it.
Heslop-Harrsion ran regular field trips to the then privately owned Hebridean Island of Rum, off Scotland’s west coast. He was the only academic to have permission to do so from the land-owners and the Island was off-limits to all others. It was during these trips that he, and his students, began to find species of plant that would be normally associated with southern Europe. These plant finds appeared to provide the definitive evidence to support the theory. His work was published in leading journals and the plants went on to be officially included in the British Flora. It was a good example of an idea, being tested in the field and then becoming a validated theory.
However, as more finds were made each year, some were becoming sceptical, particularly at Cambridge University. To rather unfairly cut a long but intriguing story short, the aforementioned John Raven managed, with some degree of subterfuge, to ‘negotiate’ a place on one of the field trips. Unknown to Heslop-Harrsion, he had gone to verify the finds.
After a few days on the site, a new ‘southern’ plant was indeed discovered. Raven visited the ‘find site’ a day or so later and did indeed find the same plant growing in the location. He dug one of the specimens up to authenticate its identification later when back in the lab back at Cambridge.
On his return, the plant was indeed the described species, but then he noticed something strange about the soil around the root bole. It appeared to be different from the rest of the soil in the bag. He sent it to a pedologist (soil scientist) colleague who ran an analysis of the soil around the root. He found the presence of specific minerals that could not be associated with the Island of Rum, they simply did not occur there. When they checked, these particular minerals almost had a geographical ‘finger print’ in that were nearly exclusively found in the soils of North East England, more spcifically, around Newcastle. The only conclusion was that they had been grown in Newcastle, then transported and planted on Rum.
Of course this is a summation of a long book and all sorts of twists and turns ensued including the status of Raven as an amateur botanist, against Heslop-Harrsion’s professorship, the role and status of the two institutions, Raven’s subterfuge etc. It does make a fascinating read.
However, the point here is what makes an emeritus professor, in a leading department, at a top university, fabricate data? Why should anybody at any level do that? Certainly, today academics are under terrific pressure to publish. I noticed recently that Peter Higgs (he of the Higgs boson) said that he’d never get a job now as he’d only published a couple of papers in his career. However, making up data to fit the theory remains a curious step to take. Interestingly, the very limited research into this area, suggests that once somebody does it, even at a fairly early career stage, they continue to do so later in their careers! Just cutting the odd corner, let’s just say that n=100, rather than n=10 (the mean won’t shift much). We repeated the process how many times? Let’s not include those data as they throw the mean out, or if we ignore those data we get a significant result, etc, etc, etc.
There was some work a few years ago that suggested a link between academic dishonesty at University and unethical behaviour in later years. If we want science to ‘do what it says on the tin’ in relation to understanding the world the world around us, what’s the point of making stuff up? If it’s just to get on with some sort of career, well, that might explain some of the unethical behaviours we’ve seen over the years.
At the end of the tutorial I said to the students, “Another thing, don’t come here next week and say “my experiment didn’t work”, experiments only ‘don’t work’ when you don’t get any data. If your results don’t fit the theory, well that’s a good thing isn’t it? It challenges orthodoxy, questions the established, confronts convention.”
They looked back at me with the sort of resigned expressions that showed they were thinking they wished they’d got someone else as supervisor. “Whatever you say” said one of them spectacularly missing the point.
Sabbagh, Karl (1999) A Rum Affair how botany’s ‘piltdown man’ was unmasked. Allen Lane, London.