McGilchrist opens this chapter with these two aptly chosen quotes which pinpoint the key arguments he is making in this chapter.
‘Most human institutions, by the purely technical and professional manner in which they come to be administered, end by becoming obstacles to the very purposes which their founders had in view.’ (William James, 1909, A Pluralistic Universe)
‘Specialization is for insects.’ (Robert Heinlein, 1973, Time Enough for Love)
This chapter is a bit different to others In The Matter With Things as it focusses less on hemisphere correlates and is not heavily reliant on the hemisphere hypothesis. Instead McGilchrist explores the question of whether we can trust science’s claims on truth and examines the limitations of the institution of science.
I recognised most of the key points McGilchrist makes in this chapter as I think would anyone who has a background in research and publication, especially but not solely if these are related to science. Many of the same issues arise in humanities disciplines. McGilchrist discusses the limitations of the institutions of science under four headings; Specialisation and its impact on original thinking; How reliable is scientific evidence? The problems of publication; and Peer review.
Specialisation and its impact on original thinking.
‘Science is a victim of its own extraordinary success’. (p.502) The explosion of scientific knowledge has led to increasing specialisation, such that scientists can only be an expert in a small area. When even a very good scientist talks about science, unless he is talking about his own area, he is taking it on trust/authority. We need to think about the worthiness of this trust/authority.
Specialisation drives scientific disciplines apart, leads to ‘narrowness, technicalisation and fragmentation, at the expense of breadth, humanity, and synthesis’ (p.508). McGilchrist argues that whilst of course we need specialists, we also need generalists. We need both the flies eye view and the birds eye view. He describes this as follows:
‘If perceiving shapes is how maths and science progress, as I believe it is, you will see those shapes only by rising above the hole where you are digging. The view in the valley floor is good, but if you never climb, you will not know that there are many other valleys, and mountain ranges nearby, which are not only beautiful in themselves, but help you see why good work needs to be done down in the valley floor at all.’ (p.504)
Specialisation also leads to specialised jargon.
‘Increasingly, the heavily acronymic jargon of research papers seems to me to present an almost impenetrable barrier to anyone other than the most highly specialised reader, and even then, if they are to get anything out of the exercise, they must have a huge capacity to tolerate boredom’. (p.507)
(McGilchrist’s writing often makes me smile 😊)
How reliable is scientific evidence?
In this section McGilchrist discusses the problems involved in interpreting data, taking mirror imaging (a way of knowing what is going on in the minds and brains of people) as an example. Brain activity scans are difficult to read accurately. The data require interpretation and therefore cannot be assumed to be objective or truthful. Every way of looking at the brain has its limitations. Whenever you are looking at a complex system, you can’t assume that the bit you are looking at is the crucial one. We should use as many ways as possible to look at the brain, and not rely solely on scanning.
In scientific research, for a result to count as important it must be replicable, reliable, and reproducible, but ‘A survey of 1,576 researchers across scientific disciplines published in Nature revealed that more than 70% of researchers had tried and failed to reproduce another scientist’s experiments, and more than half had failed to reproduce their own experiment.’ (513)
A widely cited paper by John Ioannidis (2005) – ‘Why most published research findings are false’, concludes that most research is not adequately designed to prove what it claims to show, that ‘The hotter the scientific claim, the less likely the research findings are to be true’, and that ‘the greater the financial and other interests and prejudices in a scientific field the less likely the research findings are to be true’.
There are now huge temptations (in terms of financial and reputational rewards) to commit anything from a minor misdemeanour to recognisable fraud (fabrication of results), and McGilchrist provides examples of these in this chapter.
McGilchrist also includes an Appendix (3) on the reliability of public health policy which makes for interesting reading. If you have cut salt out of your diet, you might want to think again, or read the Appendix!
The problems of publication
Most academics will recognise the exhortation to Publish or Perish!
Institutions put enormous pressure on their staff to publish, whether or not they have anything to say; quantity is more important than quality, as is publishing in high impact journals. This leads to corner cutting and inflation of claims. It also leads to a focus on writing short papers rather than books, which take a long time to write and require fallow periods. McGilchrist’s view is that this is ‘inimical to free thinking.
‘Scientific thinking gets crystallised too early, before it has had a chance to broaden and deepen; there is no longer a chance for ideas to evolve, to enter the necessary fallow period of unconscious gestation, without being prematurely forced into explicit form, and worse still in sliced form, so that what might have come to be a dawning new Gestalt is forever lost. And in the end, science is not about producing data so much as thinking, to which the acquisition of data can be only a prelude or addendum.’ (p. 516/517)
This pressure to publish can also lead to deliberate gaming of the system, where authors chase citations by working in ‘highly populated’ areas of science (even though ‘it is estimated that only 20% of cited papers have actually been read’, p.158), or even pay to have their work published in predatory open journals (See Beall’s list ) As soon as there is payment for publication, the whole system is corrupted.
And then, there are the fake papers. McGilchrist devotes Appendix 2 to some of these – papers such as those that are created by computer programs but nevertheless succeed in getting published, despite being, literally, gibberish.
So, we may ask, what happened to peer review?
Peer review
How effective is peer review? Richard Smith (Editor of the BMJ) wrote that far from being an objective, reliable and consistent process, peer review is ‘a subjective and, therefore, inconsistent process ….. something of a lottery’. (Smith, 2006, ‘Peer review: a flawed process at the heart of science and journals). As McGilchrist states:
‘Bias is intrinsic to human life. We just waste a lot of time and money pretending we’re avoiding it, and then kid ourselves that the outcome was ‘objective’ – a more dangerous position, because it introduces complacency and is a much more difficult thing to fight, precisely because of its appearance of objectivity’. (p.529)
Peer review is a laborious progress which takes up researchers’ time, which is given for free, and so takes them away from their own work. Interestingly, McGilchrist tells us that until the 1930s/40s peer review was never part of the publication process. Papers were reviewed by the editorial committee. Einstein, for example, refused to subject his work to peer review – only one of his 310 publications underwent peer review. Presumably once was enough to convince him of the flaws in the process.
There is also evidence that peer review can be prone to bias against innovation and radical new ideas, such that no-one wants to publish a paper that will rock the boat. Those who step out of line pay a huge price. In addition, reviewers have been shown to regularly fail to spot major errors in research, such that the process is obviously open to fraud.
The bottom line is that science is not exempt from human fallibility.
McGilchrist ends this chapter by discussing the need for a new paradigm, one that recognises that the essence of good science is originality and original thinking takes time. Science cannot avoid operating under the existing paradigm, ‘because, without such a paradigm, its findings could not cohere’ (p.536) but working within the prevailing paradigm also ‘militates against those great insights that change the direction of scientific history, despite this being widely believed to be precisely what science is about.’ (p.536)
McGilchrist believes that contemporary science is not scientific enough in that it is not willing to be aware of its limitations. On the concluding page of this chapter, McGilchrist defends science in the following terms:
‘Science is, or should be, a source of wonder that opens out our understanding of the world and gives us one of the touchstones on the path towards truth. Just because science cannot answer all our questions does not mean that it is not the very best way to answer some of them, and a helpful contributor to answering many more. And that there is corrupt practice in science does not make it different from any other human enterprise.’ (p.544)
For discussion of this chapter between Iain McGilchrist and Alex Gomez-Marin, see
References
McGilchrist, I. (2021). The Matter With Things. Our Brains, Our Delusions, and the Unmaking of the World. Perspectiva Press.