How much should societal considerations influence scientific research? Striking the right balance is crucial.
//

Limiting extra-scientific influences in science

How much should societal considerations influence scientific research? Striking the right balance is crucial.

Science, as a human activity, seeks to gain knowledge about the world and ourselves. Although it obeys strict rules ensuring the truth and objectivity of such knowledge, it is also embedded in the wider society. Society can influence science through what philosophers call extra-scientific values — considerations and preferences beyond strict scientific criteria, such as political, moral, or economic considerations.

While extra-scientific values can enhance the social responsibility and democratic nature of science, they also threaten its objectivity. This is why they have traditionally been excluded from the “core” stage of scientific reasoning, where hypotheses are accepted or rejected (Figure 1). These societal values may influence various phases of research: the “upstream” phase, where research avenues and questions are chosen; the “downstream” phase, where results are communicated and used; and the “parallel” phase dealing with organisational aspects of research. However, they should not, or only limitedly (see below), influence scientific knowledge, in conformity with the fact/value distinction. It is crucial to recognise that our preferences do not determine the truthfulness of scientific findings.

However, in recent decades, there has been a growing tendency in the philosophy of science to suggest that values do and should influence all stages of scientific research, including, in large and various ways, the core phase. But both of these claims seem at odds with scientific practice.

Figure 1. The phases of scientific research
Credit. Author

Assessing the descriptive claim that values do influence science

Some philosophers of science argue that extra-scientific values inevitably influence scientific reasoning, suggesting that it is practically impossible to conduct science without them. For instance, Heather Douglas contends that while scientific values, like simplicity and consistency, tell us how strong the evidence for a hypothesis is, they do not tell us if it is strong enough to accept the hypothesis.

According to Douglas, only extra-scientific values, tied to the social consequences of making an error in accepting or rejecting the hypothesis, can do this job. But such a claim raises doubts. It seems impractical to consider such values every time a scientific claim is made, potentially rendering scientific inquiry unworkable. It seems more plausible that, in most cases, scientists adhere to their own intra-scientific standards of evidence. In any case, claims such as Douglas’s need to be empirically assessed beyond a few case studies to be generalised.

Evaluating the normative claim that values should influence science

If extra-scientific values influence scientific reasoning, they threaten the truth and objectivity of scientific knowledge. Consequently, this can have adverse effects both within and outside science. Within science, further research may be built on false claims. Outside science, scientific knowledge, often regarded as our most reliable knowledge, is used for all practical applications, including critical ones related to safety or health (Figure 2).

Suppose we allow extra-scientific considerations, such as the precautionary principle, to influence our decisions regarding hypothesis acceptance. In that case, it is advisable to distinguish between the scientific corpus – which represents our multi-purpose knowledge and should be subject to rigorous acceptance requirements – and the hypotheses forming the basis for theoretical or practical decision-making (for example, if we want to base future research or formulate policy measures based on an uncertain hypothesis).

Figure 2. Extra-scientific values can have harmful influences, such as promoting an inefficient drug.
Credit. World Health Organization, CC BY-SA 3.0 IGO.

Three requirements for value influence

Therefore, any incorporation of extra-scientific values in science must adhere to the following requirements:

Any model for values in science must: (1) ensure the truth and objectivity of scientific knowledge; (2) clearly state the uncertainties associated with scientific knowledge; (3) distinguish between scientific knowledge and claims taken as a basis for action.

Philippe Stamenkovic

Implementation

Several models in the philosophy of science align with these requirements. For one, Sven Ove Hansson’s corpus model only allows extra-scientific values to influence scientific reasoning if they strengthen the level of evidence required to accept a claim (in other words, only if they reinforce the truth of scientific knowledge). Gregor Betz emphasises the need to clearly state the uncertainties associated with scientific claims instead of bridging them with values.

Moreover, these requirements resonate with scientists’ own cautiousness and preference for false negatives over false positives. Methodological documents from regulatory and intergovernmental agencies and institutions such as the US Environmental Protection Agency, the European Food Safety Agency, and the International Agency for Research on Cancer, among others, also illustrate the requirements. Therefore, these requirements seem to align with scientific and expert practices.

Three ways in which extra-scientific values should influence science

Science is a specialised activity aimed at objectivity and truth, but it is also embedded within the wider society. While extra-scientific values can enhance the social responsibility and democratic nature of science, they also threaten its objectivity. Therefore, great care must be taken when letting these values influence scientific research:

  1. Extra-scientific values can and should play a role in shaping the “upstream” phase of research, where research priorities and questions are set. Society should contribute to these decisions, ensuring that scientific endeavors address social needs and utilize public resources responsibly.
  2. Extra-scientific values can and should guide the “downstream” phase of scientific research, where results are communicated and applied. Scientific knowledge should not have negative social consequences.
  3. Extra-scientific values may influence the “core” phase of scientific research, where hypotheses are tested, only if they strengthen the truth and objectivity of scientific knowledge. Otherwise, their consequences can have adverse effects on both science and society.

🔬🧫🧪🔍🤓👩‍🔬🦠🔭📚

Journal reference

Stamenkovic, P. (2024). Straightening the ‘value-laden turn’: minimising the influence of extra-scientific values in science. Synthese203(1), 20. https://doi.org/10.1007/s11229-023-04446-2

A former engineer, Philippe Stamenkovic is a Marie Skłodowska-Curie postdoctoral fellow at the Department of Philosophy of Uppsala University (Sweden). His research interests include socially relevant philosophy of science and the relationship between science and society.