The Common Themes and Diverse Practices of Scientific Methods

Introduction

Science in some form or another has been practiced for millennia, to varying degrees of success, and throughout this time, people have argued about which is the way to practice science. The study of the scientific method can be described as the attempt to work out how success is achieved in science. The definition of success in science is a challenging topic in itself; however for the purposes of this essay, we will take it to be developing, testing and accepting theories which are closer to the truth (explain more observations etc.) (Niiniluoto, 2019). The debate over scientific method climaxed in the 20th century, when the unificationist ideas of Popper and the logical positivists were challenged by Thomas Kuhn when he introduced the concept of multiple scientific methods (Andersen & Hepburn, 2016). This concept was later supported by the radical ideas of Feyerabend, which led to an increased acceptance of there being multiple scientific methods. This can be clearly seen in the development and specialisation of the field, with philosophers of science now choosing to study a single scientific field instead of science in general. For this reason, and many others that I will discuss through the analysis of method between times, fields and individuals, there are multiple scientific methods, all of which are valid and linked by a set of core values, which vary in their specifics and importance between methods.

Unpacking the Components of Scientific Inquiry

Multiple scientific methods do exist, but they all share a very similar set of core values. Any method involves observation, hypotheses, experimentation, analysis and appraisal, which makes it easy to assume that there is a single scientific method, the one commonly presented in science education. These steps were first introduced by philosopher John Dewey and Karl Pearson, a mathematician (Andersen & Hepburn, 2016). They seem convincing at first, however after further thought, it is clear that this is not just a method of science, it is a general method of enquiry. This makes sense, as science education, especially at school, is designed to teach students how to be inquisitive and solve problems, not just do science. Another reason for the prevalence of this apparently ‘general scientific method’ is in order to make the matter of science more accessible to the general public. After considering the existence of multiple scientific methods, Weinberg reflected that the consequent lack of clear method “raises problems for the public understanding of science [as there is no] fixed scientific method to rally around and defend” (Weinberg, 1995). This explains the persistent belief that these features underlie science and therefore constitute a method (in fact, the existence of common criteria in scientific practice convinced unificationists such as Popper, Mill and Carnap that there must be a single, general methodology (Tianji, 1985)), however this is not the case. They are simply components of most scientific methods and can differ in their exact use and importance between specific methods. For example, the exact way in which experiments are done can differ, and the distinction between them is not clear-cut. The two extremes tend to be described as theory-driven, in which experiments are done with the sole purpose of testing a hypothesis (deductive), and exploratory, where data is just being collected randomly in the hope that a new theory will be developed from it or a current theory will be supported by it (inductive). So already, a scientist has two possible ways in which to carry out experiments, but they are not distinct. Theory-driven experiments can involve exploratory features, such as collecting large amounts of data to determine parameters which fit the theory (for example the mass collection of data at the moment in order to create a model for the spread of COVID-19), and exploratory experiments can never truly be theory-free (Andersen & Hepburn, 2016). There must be some kind of background knowledge to know what to measure or look for. For instance, it could be argued that the Hubble Telescope was purely exploratory, with the purpose of simply exploring the universe, yet the scientists must have some idea of what to look at in order to make any worthwhile observations. As such, there is a broad range of ways in which just experiments can be done, let alone the various methods of analysis, appraisal and observation. So, all scientific methods follow common themes, but each theme can differ significantly in its specifics, so therefore the overall methods are not the same.

As mentioned previously, science has been practiced for millennia, but the science done by the likes of Aristotle and Archimedes is very different to what we view as science now. This is largely due to the fact that their knowledge of the world and the resources available to them were severely limited in comparison to today. As such, their approaches when finding out more about the world differed. For example, the content and methods of physics have changed drastically in the past centuries. The two main changes came in the shift towards the Newtonian view of the world (mechanisation and an increase in the importance of mathematics) and after Einstein’s introduction of the worlds of relativity and quantum mechanics. In both these cases, a new way of looking at the world was introduced, and as such new methods of probing it were also made possible. In addition, the concepts of what was true changed. After the move into Newtonian physics, people realised that you didn’t just have to observe- good mathematics could give much more certainty and precision to descriptions of the world. Once Einstein’s revolution of physics was over, physics moved away from explaining the observable phenomena and towards the abstract notions of sub-atomic particles, resulting in a massive change in approach and equipment, and as such, method. So, over time new discoveries and ways of viewing the world changed the way in which science was done, but at the same time did not completely overthrow the scientific facts developed beforehand (Lee, 1943). This has happened numerous times, both explosively as seen in the previous examples and gradually. As such, there have been multiple scientific methods throughout history, all of which have been successful in their own right. In addition to new knowledge, new technology has also shifted scientific methodology. When Galileo invented the telescope, new, more precise ways of exploring the universe were created, allowing for clearer methods and descriptions. In more recent times, the use of computers which are able to collect and process huge amounts of data have shifted scientific practice to a more inductive method, where patterns in data can be found rapidly and hypotheses developed from them (you could say that this means that Bacon’s ideal method is being realised). Computers have not just revolutionised the way we form hypotheses and perform experiments, they have also made tests more theory-laden, meaning validation of the data and hypotheses that result is much more challenging (Andersen& Hepburn, 2016). As such, new methods of appraisal and validation have had to be used, leading to a new scientific method. So, scientific methods evolve through time, thanks to new theories and instruments leading to a need for new approaches.

Scientific methods don’t just differ throughout time, they also differ from field to field. This is largely due to a different prioritisation of values, which will inevitably lead to considerably different methods. For example, biomedical scientists don’t necessarily think that knowing absolutely everything is important, they focus on things that are relevant and have the potential to save life. For instance, in pharmacology, every molecule is not tested on every receptor to find out if it has a slight effect, instead specific receptors relevant to a disease that needs treating are chosen, and intelligent choices in drugs to test are made. In addition, on many occasions, drugs are found and used without a complete understanding of their mechanism of action. In other words, the need for overall, complete knowledge is much less of a priority for pharmacologists, their primary goal is to improve health. In contrast, theoretical physicists don’t focus on the practical utility of their discoveries, and instead their main motivation is to find out as much about matter and the universe in as much detail as they can. So, whilst they are still systematic in their approach, they continue investigating one thing for years, even if an initial solution is found. They simply explore further, making their methods of doing science very different to those used by pharmacologists. Given there is no argument on the fact that both these fields constitute as science, that they are both successful in their own right, and they still follow the general values outlined earlier, there is no way of claiming that there is only one scientific method, as their methods differ greatly. These differences in method between fields led to and are highlighted by the benefits of inter-disciplinary work, and the important progress that often results from such collaborations (Spiece & Colosi, 2000). An example of this is the huge advancement in the field of molecular biology in the first half of the 20th century. The area was largely stagnant until physicists and chemists such as Linus Pauling and Max Delbruck began to be interested in the field. They brought new ways of approaching the analysis of molecules in the body, as well as new instrumentation. In other words, they introduced an entirely new method. Once this was widely adopted, the field exploded and led to some of the most influential discoveries of the last century such as the discovery of the structure of DNA and the cause of Sickle Cell Anaemia. Before the involvement of the physical scientists, the field had still been progressing, in fact it led to the development of genetics and other key biological disciplines. So, their initial method wasn’t necessarily wrong, and as such was a valid scientific method. The influence of the physical scientists simply demonstrates that differences in method do exist between disciplines, else the physical scientists wouldn’t have had much of an impact on the progress of the field. As such, there is no single scientific method, as different fields have different priorities and therefore different approaches.

The Social and Personal Influences on Scientific Method

Whilst it is quite easy to see that there are differences in scientific method throughout history and between disciplines, identifying differing methodologies between scientists in the same field is more challenging, however such differences do exist. The reasoning behind such variation in method was introduced by Kuhn, and can in fact be used to explain the differences in method discussed above. Kuhn claims that the act of observation does not limit scientific beliefs to just one, it simply creates a short list of possible scientific beliefs of which scientists must decide to follow one. The belief/theory you choose depends heavily on historical and personal factors. In other words, scientific decisions and method are closely linked to social and personal influences, which leads to variation. Once you choose a certain scientific belief, this affects the method you take. Philosophers Putnam and Shapere highlight this in their work, claiming that as the context of science changes, so does the method, with Putnam coming to the conclusion that the only way in which a “formalisable method” could be obtained is by it “being isolated from actual human judgements about the contents of science”, and that this is essentially impossible (Tianji, 1985). The differences in personal beliefs and outlooks will lead to different prioritisation and interpretation of each of the core values discussed above, resulting in a different method. As such, there is no one scientific method, and each scientist, with their own personal backgrounds and beliefs will have their own distinct method.

This is not to say that the methods used are all alien to one another, they all share a similar set of values, as discussed earlier, but the different approaches are obvious in the history of science. For example, Rosalind Franklin and Watson and Crick had the same goal: to uncover the structure of DNA. Their approaches, however, were completely different. Franklin, with her background in X-ray crystallography was determined to use only data obtained from this technique to discover the exact structure, and she got very close, uncovering the key bit of information required for Watson and Crick to be successful in the determination of DNA’s structure. The reason they got there first was because they used a different method. Instead of just looking at data and trying to glean information from it, they developed numerous models, and saw if they fit the data. Their final, correct model was a physical one, and perhaps this more hands-on, visual approach was the reason behind their success. Both methods employed here had the same underlying features; experimentation, hypothesis formation and testing, but they were clearly different, proving that there are multiple viable scientific methods. A further, more extreme example of this is the different, successful approaches taken to develop drugs in Pharmacology. In the development of antibiotics, there were two key scientists, who created highly successful drugs in very different ways. Paul Ehrlich used a targeted, systematic approach to develop synthetic sulfa drugs, whilst Alexander Flemming made the serendipitous discovery of penicillin. Both these drugs were widely used for decades, showing that within the same field, varying methods can be used, each to great success. These different methods don’t just hold for theory development, they are also seen in theory appraisal and decision making. This is seen on a regular basis in all areas of science, where a contradictory piece of evidence appears for a theory (Spiece & Colosi, 200). Some scientists will choose to develop an entirely new theory (akin to Popper’s suggestions in his falsification theory), whilst another will simply adjust the current one (as described in Lakatos’ research programmes). The use of both these methods is common, and doesn’t prevent science from progressing, yet they are very distinct, demonstrating that differing scientific methods do exist and are key to the practice and development of science.

It has now been made clear that between time periods, disciplines and even individuals, there are differences in scientific methods used, and as such there is no single, unifying scientific method. Yet in the current scientific world, it is often difficult to see such a variation in method. In science literature, papers almost always follow the same narrative; a neat, linear representation of the work they did, making it seem that there is a single common method (Andersen & Hepburn, 2016). However, in reality the scientists involved in the papers differ greatly in their methods, and significant portions of the work that led them to the paper, including many failed attempts, are never mentioned. This was highlighted by Sir Peter Medawar in his talk Is the Scientific Paper a Fraud?, where he outlined how all papers follow the same structure (Introduction, Methods, Discussion, Conclusion), which does not mirror the actual process of science, in which much of the discussion comes before any kind of experimentation (Medawar, 1964). The discussion is where much of the variation appears (different choices in approach), and when put after the results, this variety is largely hidden, in turn removing the display of multiple scientific methods. This appearance of a single scientific method is, as already discussed, caused in part by the nature of science communication through papers, which are simply retrospective reconstructions of the work truly done, but also by the nature of the struggle for funding in the scientific community today. Funding bodies need to have a level of certainty that the work they put money into will bring about useful and relatively reliable discoveries. As such, they are unlikely to fund projects involving unorthodox methodologies as a result of their unreliable nature. This means that scientists are often forced into a particular way of doing science just to be able to get funding, regardless of their own instincts and wishes. Therefore, it may appear to the public that there is a single way of doing science, but, as I have demonstrated above, there is not.

Conclusion

In conclusion, scientific method is fluid, and despite the huge amount of research put into it over the last century, there is no single scientific method. Methods vary between times, fields and individuals due to a difference in prioritisation of the values which underpin all scientific practice. These differences often come about due to changes in the social or political climate, which create norms, which can potentially lead to the appearance of there being only one scientific method. This is seen more than ever in modern times, when scientists have significantly less freedom than they have in the past as a result of the constraints imposed by funding pressures. Even when methods do differ, scientific papers, often the sole report of research, make it seem as if all projects follow the same pattern, which is not the case. There is still massive variation in scientific method, and this makes it the dynamic and exciting field it is.