A few blogs ago, I quoted the philosopher Ludwig Wittgenstein who wrote: ‘One always forgets the expression “I thought I knew”.’ Wittgenstein is speaking colloquially here, ie., he is referring to how we use “knew” in common usage. But the same question of what we mean by knowledge and knowing applies to science in cases when an established truth is overturned. And it is to how scientific progress is made when new truths are established that I’ll turn my attention to in the work of Thomas Kuhn.
Thomas Kuhn was a philosopher of science who published The Structure of Scientific Revolutions² focus was the dynamic or underlying structure of how scientific discoveries are made. Prior to him, most people thought science was a relatively straightforward matter of accruing new facts. Scientific discovery was thought to consist of discoveries that expanded what we knew. Under this view, the difference between science in the 21st century and science in the 17th century is that scientists in the 21st century know a lot more about the world. As our knowledge accrues, our scientific theories increasingly approximate the truth. Also in this view, great scientists might enable us to make more rapid progress in new discoveries or new domains of knowledge.
Of course, Kuhn doesn’t contest that we know more know than we did then. That’s not his point. What he does claim is that it isn’t the amount of knowledge that categorizes scientific progress. He does describe scientific progress as being characterized by episodic alternations between phases he calls “normal” and “revolutionary”. These phases are categorized by a consensus or model of how the world is understood conceptually in a period of normal science only for that consensus to dissolve in the period prior to the emergence of a new model, the revolutionary period. The picture that emerges looks like this:
- This cycle is preceded by a pre-paradigmatic phase characteristic of an immature science: there is no consensus on how the observations fit together and no model to define directives for future research.
- The period of normal science is one that has a broad consensus of the scientific community, a model or paradigm that explains many scientific observations and a map towards what future research would look like.
- The period of model drift stems from the success of the paradigm. The paradigm guides research and in doing so new observations are made that cannot be explained by the existing model. These observations are typically waved away as anomalies by the scientific community who maintain confidence in the existing paradigm.
- The period of model crisis is one where many in the scientific community can no longer reconcile the existing model to the increasing number and seriousness of the discrepancies of new observations from what the model predicts.
- A model revolution marks the emergence of a new model that explains the new observations that were left unexplained by the previous paradigm. This new model is often not accepted by many in the scientific community who for a while still attempt to explain away the observations that were problematic for the old model.
¹ Photo Taken from the New Atlantis: https://www.thenewatlantis.com/publications/did-thomas-kuhn-kill-truth
² Kuhn, Thomas. “The Structure of Scientific Revolutions.” https://www.lri.fr/~mbl/Stanford/CS477/papers/Kuhn-SSR-2ndEd.pdf
³ Interestingly, Kuhn’s work popularized the usage of the term “paradigm”. The word was used as early as the 15th century in Latin but it was Kuhn who gave it the philosophical specificity of a coherent structure from which to interpret scientific observations. Since then, the word has been applied far and wide in virtually every type of human activity from supply chain management to new age spirituality. So much so, that as I searched for an image to encapsulate Kuhn’s work I was led to many web pages that were anything but scientific journals. Because of this, in his later work, Kuhn walked back his usage of the term paradigm and began using “disciplinary matrix”. Nonetheless, the term paradigm is in wide enough usage that most people have a general if vague understanding of what the term connotes. The particular image above was taken from an online article called “Advancing Empirical Science in Operations Management Research: A Clarion Call To Action”. https://pubsonline.informs.org/doi/pdf/10.1287/msom.2019.0829