Is science sometimes in danger of getting tunnel vision? Recently published ebook author, Ian Miller, looks at other possible theories arising from data that we think we understand. Can looking problems in a different light give scientists a different perspective?

A Chem ArXiV?

The recent edition of Chemistry World had an article in which it was reported that the American Chemical Society is planning to start up a ChemRxiv, a preprint server. In this context, I note that chemists are strangely resistant to change, and the article was hardly overly enthusiastic about the concept. You probably know that the physics community has preprints on the ArXiV server, and by and large, physicists submit papers there for community peer review before submitting to a final journal. I started writing an article for this blog a little while ago on this topic, and never finished it, but this part of what I wrote. This, to me, makes a lot of sense, and a number of years ago some prominent chemists started up the Chemistry Preprint Server. The response from chemists was strangely disappointing. Very few submitted anything, and of those submitted, some were distinctly of low quality. Nevertheless, it seemed a good idea at the time. However, the main part of the chemistry community then proceeded to ignore it. Now I feel this should have been a great place for two things: getting an important paper into a form that the average community would accept, and secondly getting enough support for it that a journal would publish it. However, that preprint server did not work because very few were interested.
According to the Chemistry World article, the previous preprint server died because the quality of the publications was too low, and because the American Chemical Society would not publish anything that had been published before. That may be unfair on the ACS; I doubt it was unique in that. My belief is it died because most of the chemists of stature refused to take part, although in fairness at the time I also refused to put my experimental work on it because I was afraid that it would not be accepted later by the journal where most of my experimental work was then being published.
What are the purposes of a preprint server? One might be to get information out there earlier, but my feeling is that is not really that important in most cases. What I used the earlier server for was to archive material that was otherwise difficult to publish, and possibly get some idea what I could do to improve its acceptability.  Now, you might say, reading that, that I was a contributor to the "low quality" chemistry. Before jumping to such conclusions, consider these.
I submitted a logic analysis once to a chemical journal, the purpose was to show that the published data did not support the concept that the cyclopropane ring could delocalize electrons into adjacent unsaturated substituents. The standard position was it did, largely because it stabilized adjacent positive charge and gave bathochromic shifts to many UV absorptions, and extending the wave function would do that, as in the allyl cation. However, there is another way, namely a polarization field, i.e. the problem is one of electromagnetic theory rather than quantum mechanics. I wrote a logic analysis that showed not only did this alternative give correct predictions, but there were over 60 different types of experiment that falsified, or at least cast a lot of doubt, over the then accepted "delocalization" explanation. Note that demonstrating that cyclopropane stabilized adjacent positive charge 250 times still only adds one datum into a logic analysis. As a further aside, one previous review argued that quantum computations verified the presence of extra stability through delocalization. You may be amused to note that such computations used the same MO programs that "proved" the stability of polywater.
The first journal rejected it because they argued "it was not what the average chemist wanted to see" (or something like that). There were not enough diagrams (I assumed that a triangle headed towards a substituent X should have been general enough for many cases). One journal rejected the proposal "because this issue is settled". Who cares about the data and the logic? Others simply said they did not publish logic analyses. Fair enough in one way, but how could it get published? I will concede immediately that it could have been made easier to follow, but what I hoped was that chemists would tell me what it should have looked like to convince them. No such luck.
Similarly with a paper where I showed a pencil and paper route to getting a reasonably accurate estimate of the covalent bonds of the group 1 elements. That was rejected by editors because "nobody would be interested". They may or may not have been right, but the interesting point is it argued there was a hitherto unrecognized quantum effect applying.
Now some will say that was rubbish. Either it was right or it was wrong. If it were right, should it not be published somewhere? If it were wrong, should not someone be able to show where? And that is where a preprint server should be useful. It gives the scientific community the opportunity to comment and act as public referees, and takes away the ability to block something by editors who do not have to give excuses.
Posted by Ian Miller on Nov 3, 2016 9:44 PM Europe/London

Share this |

Share to Facebook Share to Twitter Share to Linked More...

Leave a comment?

You must be signed in to leave a comment on MyRSC blogs.

Register free for an account at