Editor's note: In last Thursday's programme Melvyn Bragg and his guests discussed The Scientific Method. As always the programme is available to listen to online or to download and keep - PM.


    After the programme Simon Schaffer made a dash for King's Cross, the train to Cambridge and what looked like a full day's lecturing. He talked about our New Year series, The Written World, in which he played a magnificent role explaining the value of Newton's notebooks.

    It seemed that the programme arrived at a time when the Cambridge Library was about to open up access online to some of the extraordinary written material it has, and so was an accidental bull's-eye. Michela Massimi said that the experience had been different from what she'd imagined. It was 'cosy'. And instead of talking to lots and lots of people, she realised she was only talking to one or two. Once again, someone who came on to the programme, never having been there before, and within five minutes of being in the studio was thrown into the deep end, simply sailed after an initial shyness.

    John Worrall, who neatly reversed the two questions I wanted to ask him (deduction should always have preceded induction of course!), went into a wonderful aria about the need to do a programme on evidence-based medicine.

    He spoke powerfully about the effect of placebos, saying that you could cut somebody's ribcage open after an attack of angina, and instead of operating just stitch it up again and say you had done the usual surgery, and that placebo effect could result in just as good a recovery as that resulting from the surgical intervention.

    He put it more elaborately than that, but certainly he gave both Tom and me the taste for evidence-based medicine in a future programme.

    Off across the road to a meeting with a writer from Caldbeck in Cumberland, Kathleen Jones, who is writing the biography of Norman Nicholson, an extraordinarily fine Cumbrian poet who died only a few years ago. He was on the favour and favour list. He won the Queen's Medal for Poetry. He was highly praised by no less a critic than the late Ian Hamilton. At the moment his poems are out of print, but I hope this book will encourage people to put him back in print.

    And so to the office for a meeting about the new South Bank Show, looming in May, and then to lunch with an old friend, and down to the House of Lords to listen to some of the debate on the Scotland Bill; and meet a young man who is, as it were, a secular godson and have tea, a wander round and a visit to the Commons, where the British Armed Forces were being massively defended from the Tory benches and the whole question of cutting them would signal the end of the known moral world.

    And no park today, but just the London streets in the most extraordinary winter light. Such bright blue skies and such a chill tang in the air.

    Maybe the air is not fresher when it is cold and tangy like this, but it feels so much fresher. Pavements up and down Whitehall crowded, of course, with school parties, but the good weather, as usual, brought out a feeling that we were all very lucky to be enjoying it.

    And then the evening.

    Best wishes

    Melvyn Bragg

    Tagged with:


    This entry is now closed for comments.

    • Comment number 8. Posted by John Thompson

      on 8 Feb 2012 06:01

      Frank Cioffi who died 1 January 2012 moved from an early fascination with Popper’s falsifiability test for distinguishing science from pseudo-science to a view that criticisms of Freud as a defective scientist were misguided.He saw Freud as a flawed humanist whose insights were comparable to Nietzsche’s and the great novelists of his time.Among psychologists he preferred the restlessly inquisitive William James to Freud,whose faults, he held,were grounded in failures of integrity as a humanistic investigator rather than in failures of scientific method.

      • This entry is now closed for comments. Number of positive ratings for comment 8: 0
      • This entry is now closed for comments. Number of negative ratings for comment 8: 0
    • Comment number 7. Posted by colmilne

      on 2 Feb 2012 16:18

      Re: Scientific Method.
      What if a flock of Aves ( species all white ) becomes covered in soot and is observed by a scientist, who assumes that they are a black variety ? Is this the start of myth and legend ?


      • This entry is now closed for comments. Number of positive ratings for comment 7: 0
      • This entry is now closed for comments. Number of negative ratings for comment 7: 0
    • Comment number 6. Posted by Huw Llewelyn

      on 31 Jan 2012 21:30

      The purpose of medical research is to identify symptoms and other features that in turn identify groups of patients who benefit more from having a particular treatment than by not having it. The title chosen for the theory that explains the link between the symptoms etc (e.g. rapid irregular heart rate) and the effect of treatment (e.g. digoxin) is a ‘diagnosis’ (e.g. ‘fast atrial fibrillation’). This process of identifying combinations of features and matching them to treatments requires careful thought based on tried and tested theories and if at all possible, randomised clinical trials (RCTs). It is not possible to do RCTs on patients with every possible combination of features. Therefore, a doctor has to use existing RCTs to guess what would have happened if an RCT had been done on a sub-group of patients identical in all respects to patient being advised. This is not easy. It might be improved by better communication between doctors and patients and some innovative research methods.

      • This entry is now closed for comments. Number of positive ratings for comment 6: 0
      • This entry is now closed for comments. Number of negative ratings for comment 6: 0
    • Comment number 5. Posted by John Picken

      on 31 Jan 2012 14:42

      Thank you both for these comments. In particular the medical issues stand out for me. One issue with observation is the question of what is observed in order to make a diagnosis or prediction and in studying groups what are the criteria for group membership.
      With my condition of heart arrhythmia it is very likely that the functioning and relationship of the internal parts of the central nervous system are highly significant in the development of the condition. This has been put forward by leading figures in the field.
      Yet this possibility appears rarely considered by doctors in making assessments, for example in NICE guidelines. When considering groups the composition seems almost arbitrary. Since much of the research is driven by pharmaceutical companies it often appears drug criteria based rather than linking to patient driven sensitivities or issues. Consequently as an individual the treatments often appear inappropriate. If the diagnosis does not cover key issues in the development of the condition, applies learning to a patient that derives from inappropriate group categorisations then it is not surprising that patients find themselves poorly served by the medical system. The expectation that the medical profession must know the answer encourages the patient to under value their essential contribution and the medical profession to deny that contribution when it is put forward.

      • This entry is now closed for comments. Number of positive ratings for comment 5: 0
      • This entry is now closed for comments. Number of negative ratings for comment 5: 0
    • Comment number 4. Posted by Huw Llewelyn

      on 31 Jan 2012 00:57

      Thank you for an excellent and stimulating discussion. I was struck by the close similarity between the scientific method and the ‘disciplined medical method’ used in day to day medical practice. I was therefore very interested in John Worrall’s idea of a programme on ‘evidence based medicine’. However, evidence based medicine is part of an intense process of ‘applied’ scientific method used by doctors (usually subconsciously but transparently by some). As in general science, in medicine there are (1) observations, (2) systems of predictive propositions with varying degrees of certainty (called diagnoses or hypotheses) and (3) actions (tests, treatments or experiments) to test these predictions. In day to day medical practice, the actions generate further observations and a cycle of revised diagnoses, more action leading to more evidence, and so on, as in all branches of science.

      Diagnoses, hypotheses and theories are thus systems of multiple predictions with various probabilities (usually with pithy summarizing titles e.g. ‘diabetes mellitus’ or ‘theory of relativity’). So in order to consider evidence based medicine, one must first consider the medical thought cycle of (1) evidence gathering, (2) diagnoses, (3) actions, more evidence gathering and so on, as applied to the individual patient. This is also important for those who wish for “no decision about me without me”, which is promised in the NHS reforms. ‘Evidence based medicine’ involves incorporating research done on groups of patients into the reasoning methods regarding an individual patient.

      Perhaps ‘In Our Time’ should consider this medical thought process applied to the individual patient first (as described, for example, in the Oxford Handbook of Clinical Diagnosis) before looking at ‘evidence based medicine’ based on groups of patients (e.g. placebo controlled therapeutic trials) that is used to support the medical thought process. Considering the more accessible medical thought process might also provide further insight into the ‘scientific method’ in general. For example, it would clarify the role of ‘probabilistic’ as opposed to the narrower ‘logical’ deduction and induction.

      • This entry is now closed for comments. Number of positive ratings for comment 4: 0
      • This entry is now closed for comments. Number of negative ratings for comment 4: 0
    • Comment number 3. Posted by John Thompson

      on 30 Jan 2012 22:08

      Section 3.Popper's account of scientific methodology begins from inverting the traditional inductivist method associated with (e.g.) Francis Bacon and emphasising instead the importance of science offering testable, quantitative conjectures that can be proven wrong through experiment and observation. Sounds simple enough in a nutshell but it's one of the few truly big ideas that philosophy of science has ever given us and it led Popper on to propose revolutionary (at least for philosophers) ideas about how science functions. I suspect Popper got closer than any other philosopher to capturing how science might actually work. Ironically, the anti-methodological philosophies of science proposed by noted Popper critics like Thomas Kuhn and Paul Feyerabend lend themselves much more readily to authoritarian interpretations than Popper's falsificationism ever did.Popper replaced deduction by falsification. His celebrated theory of falsifiability, arguing that the inferences made in science are not inductive but deductive; science does not start with observations and proceed to generalise them but with problems, which it attacks with bold conjectures.He criticized Wittgenstein by saying there are real problems in philosophy not illusions
      created by our language traps.Observations and experiments are theory-laden.He was a great man and swam against the stream as outsider.

      • This entry is now closed for comments. Number of positive ratings for comment 3: 0
      • This entry is now closed for comments. Number of negative ratings for comment 3: 0
    • Comment number 2. Posted by John Thompson

      on 30 Jan 2012 21:42

      Section 2.The 19th century debate between JS Mill and Whewell about how to arrive at scientific truths.Whewell favoured hypothetico-deductivism,which starts with hypotheses and tries to deduce empirical consequences we try to test.Mill’s 4 methods of experimental enquiry using induction were designed to identify the laws /causal factors governing phenomena.In a deduction we’re teasing out the consequences already in the premises.The main role of deduction is the testing process,taking the theory we deduce from it certain consequences and background material.In induction the conclusion goes beyond the premises.We’re going from the past and what is so far tested to generalizations that cover the past and future cases.In many cases we want to go beyond what we know.But in the case of white swans seen in the past,the conclusion all swans were white was false.But if conclusions have been confirmed enough,we’re are inclined to say(cf.General Theory of Relativity)they are approximately true when applied in future cases.Although Newton’sTheory lasted 2 centuries and was treated as true and science built accumulatively on this(eg the kinetic theory of heat and Maxwell’s theory of magnetism).Everything in the observable world seemed to confirm it.Einstein’s different theory challenged Newton’s.Space and Time became space-time.Observational evidence fitted both Newton’s and Einstein’s theory.Both theories are not bodies of objective truth,they are man-made hypotheses which fitted all the facts of the time.One theory can be replaced by another if it goes further in the range of its applications.Hume questioned the inductive method:there is no way of validating such methods,pure empiricism is not a basis for science;scientific laws cannot be proved and are therefore not certain.But if not certain their degree of probability is raised by each confirming instance.Kant saw the contribution of the thinking mind(conceptual)together with what there is.But science succeeds,gets results in the successful pursuit of knowledge.Einstein’s physics introduced general probabilities into the world whereas previous physics had been deterministic.Science had become fallible in its theories.Hypothesis was the starting point of a
      reasoning process.

      • This entry is now closed for comments. Number of positive ratings for comment 2: 0
      • This entry is now closed for comments. Number of negative ratings for comment 2: 0
    • Comment number 1. Posted by John Thompson

      on 30 Jan 2012 21:23

      Excellent talk again Melvyn on an important subject.Section 1Bacon: He has permanent importance as the founder of modern inductive method and the attempt at logical systematization of scientific procedure. He emphasised the importance of induction as opposed to deduction. He wanted to go beyond ‘induction by simple enumeration’ to a process of listing.Newton was impressed by Bacon to describe his own method as induction.An idea has been present since that scientific knowledge grows by accumulation,and that it grows by the use of the ‘inductive method’.At this time natural philosophers sought the truth:scientific procedure was man’s attempt to understand the material universe and how it works.For Galileo the ultimate test of a theory was to be found in nature,not in hypotheses.Newton said he did not make hypotheses. He laid down the law and derived the phenomena from it.We get rules of scientific reasoning. Bacon’s inductive method is faulty through insufficient emphasis on hypothesis. He hoped that merely orderly arrangement of data would make the right hypothesis obvious, but this is not obvious.The framing of hypotheses is the most difficult part of scientific work. The part played by deduction is greater than Bacon supposed. Often, when a hypothesis has to be tested, there is a long deductive journey from the hypothesis to some consequence that can be tested by observation. Usually the deduction is mathematical which we know he underestimated. Bacon had underrated the importance of hypothesis and theory and overrated the reliability of the senses. Karl Popper said “ The most important function of observation andreasoning, and even of intuition and imagination, is to help us in the critical examination of those bold conjectures which are the means by which we probe into the unknown”.

      • This entry is now closed for comments. Number of positive ratings for comment 1: 0
      • This entry is now closed for comments. Number of negative ratings for comment 1: 0

    More Posts