Notes on Philosophy of Science

Scott Brodie Forsyth

2 days ago (Medium.com)

Philosophy of science can be regarded as a meta-discipline, dealing with a range of meta-questions about science and scientific practice. Philosophy of science is relevant to both the natural sciences and the social sciences, where it attempts to theorise how the world can be understood and known, as well as how knowledge can be attained. At a basic level, there can be discussions of epistemology and ontology. Etymologically, ontology means the study of being, from Greek; “on” meaning “being,” and logia meaning “the study of.” Ontology describes what exists and how essentially, what the world consists of. Epistemology has its etymological origins in the Greek words “episteme” and “logos.” Episteme can be translated as knowledge or understanding, where logos can be understood as argument or reason. Epistemology thus describes the study of how the world is known or perceived. Moreover, within philosophy of science, a broad distinction can be made between naturalism and constructivism. Constructivism is based on the notion that the social sciences deal with a reality that is socially and linguistically constructed. Ontologically, constructivism relates to social phenomena and social relations that must be considered as phenomena continuously constituted via language and the social. Naturalism emphasises the idea of the sciences addressing a genuinely objective existing reality. From an ontological perspective, naturalism views social phenomena as part of the natural world. When discussing naturalism, natural laws and causality make up key factors. The goal is to pursue an explanatory scientific approach where theories are tested and understood through systematic and causal explanations.

Ontological objectivity describes how truth is independent of the opinions of subjects. An example of ontological objectivity is “Jupiter has a larger surface area than Mercury.” This statement is true regardless of what others may think about it. Epistemological objectivity describes truths such as “The Danish Parliament has 179 members,” which, for example, focuses on facts about society. These social issues are not as subjective as “what is your favourite colour,” but arise from shared beliefs and constructions. John Searle works with social facts, which can be described as epistemologically objective. From Searle’s point of view, within a certain level of social reality, causality can indeed be addressed. According to Searle, there is a kind of epistemological objectivity, suggesting that it might be possible to establish regularities. However, Searle does not advocate for unified science as such. Searle distinguishes between natural science and social science. In relation to the latter, social rules can be broken, whereas natural laws cannot be broken. Searle focuses on the everyday social reality. Thus, how people’s social life appears as natural and necessary, even though it is arguably not. Searle deals with society’s rules of the game, i.e., how these rules arise and how they can be broken down. Some of the basic concepts found within Searle’s works are collective intentionality, rules, and deontic content. Social facts are ontologically subjective but epistemologically objective. Social facts can be described as ontologically subjective because their existence is based on collective intentionality. At the same time, social facts are epistemologically objective. For example, knowledge can be gathered about whether a means of payment is considered valid. Social rules say something about knowledge, capacity, and will, whereas natural laws are exceptionless. It may happen that social rules are not followed; in contrast, natural laws will repeatedly appear without exceptions. Social rules may not be followed due to insufficient knowledge about them, or simply a lack of will to follow the social rules.

Ontologically, the world can be considered as social, consisting of social building blocks and constructions according to Searle. Epistemologically, knowledge is recognised through social facts. From a methodological standpoint, some theorists argue that it is the social reality, and justify primarily an interpretive methodology, while others advocate some form of unified science. Epistemological social constructivism is based on the idea that the social leads to ideas about the world, whereas ontological social constructivism claims that ideas create our (social) world. Searle writes about the construction of social reality, where ontological social constructivism appears, thus the social world that occurs as a result of cognition and action. Searle also notes when talking about contingency, it describes that things are different, but since things are as they are, according to Searle, it has a character, whereby predictions to the level of explanation are probably made. Contingency implies that something is historically specific and could therefore be different. Contingency cannot be described as mere chance but in relation to the given circumstances.

Logical positivism, in contrast, offers a distinct perspective on addressing scientific challenges. While social constructivism emphasises the role of cognition and action in shaping the social world, logical positivism takes a different route in tackling scientific inquiries. Logical positivism employs the verification principle as a demarcation principle. The verification principle entails that statements can be verified through observations or logical propositions; other statements must be considered unscientific. According to logical positivism, scientific practice should be structured according to the verification principle. Thus, there is a rejection of metaphysical or speculative claims such as Hegel’s world spirit. Statements other than analytic and synthetic can be considered meaningless or metaphysical from this philosophy of science standpoint, including normative statements. Logical positivism, however, can be criticised for leading towards confirmation of statements instead of examining what contradicts them. Furthermore, one cannot infer from “some” or “many” to “all”.

When we make inductive generalisations based on past observations, we are assuming that these past instances are representative of all future instances. Here, we may speak of the problem of induction, as described by Hume, in questioning whether induction leads to knowledge about future events. Inductive thinking involves the problem or assumption that past observations indicate how a phenomenon will unfold in the future. Considering induction and deduction: Induction involves drawing general conclusions from empirical data, whereas deduction begins with a theory wherein a hypothesis is formulated — a preliminary assumption about the field of investigation. Then empirical data is collected to test the hypothesis based on the empirical material. According to Hume, most of our knowledge is not based on mathematics and geometry (deduction) but on empirical observations (induction). However, the problem with observations is that necessary causal connections can be predicted or believed according to Hume. He also highlights the sceptical question of causality, stating that although a person may have an idea of causality or necessary connections, when this idea is tentatively linked to specific sense impressions, in reality, there are just several different isolated sense impressions following one another.

As mentioned, one of the greatest problems with logical positivism is its tendency to seek confirmation of statements rather than examining what contradicts them. Karl Popper sought to solve this issue. One of his significant contributions to the philosophy of science is proposing a deductive and logically valid method for inferring between theory and observations. Popper’s approach reverses the issue by initiating scientific inquiries with a falsifiable theory (T), and then conducting observations (O) to test the theory. This philosophy of science is termed critical rationalism, which has falsification as its demarcation criterion. Critical rationalism is based on formulating a theory or hypothesis in order to be able to falsify it. Scientific statements must therefore be falsifiable. Science, according to critical rationalism, appears as a continuous process. Thus, there is no final truth, but instead, the scientific approach proceeds through ongoing theorisation, empirical testing, and adjustments to refine the scientific understanding of the world.

Popper highlights falsification as a basis for science and sets up a deductive logic: if the theory (T) is true, then the observation (O) must be true. In formulating a theory (T), it may also turn out through observation (O) that the theory (T) is false. Scientific investigation thus begins with a theory (T), and then it is tested through observation (O) to determine if the theory (T) is true or false. If observations (O) are not true, the theory (T) can thus be falsified. Popper’s demarcation criterion for scientific validity is that the theory proposed must be capable of being falsified. Based on this reasoning, Popper criticises Freud’s psychoanalysis and Marx’s social analysis, as part of their theories may be difficult or impossible to falsify. A reason for the social sciences to adopt critical rationalism lies in dealing with the replication crisis, characterised by researchers facing challenges in reproducing similar results within canonical projects. This crisis often manifests when attempts to replicate the findings of initial studies fail. Additionally, critical rationalism offers a valuable framework for addressing the phenomenon of post-rationalisation, wherein unexpected statistically significant correlations are rationalised.

In formulating critical rationalism and the principle of falsification, a fundamental assumption appears in Popper that theories can be tested on an objective reality independent of the formulated theory itself. It is assumed that, based on an observation, it can be directly evaluated whether the observation supports or contradicts the formulated theory. A similar assumption appears in logical positivism, although with the significant difference that the verification criterion appears as a demarcation line. However, this assumption is challenged by the Duhem-Quine thesis, which argues that scientific hypotheses cannot be tested empirically in isolation. It is pointed out that all observations ultimately appear as theory-laden, and therefore, an ultimate verification or falsification cannot be achieved. The Duhem-Quine thesis can be said to emphasise that observations always occur within a theoretical framework. It is not possible to test a given hypothesis in isolation, as every time a hypothesis is tested, a number of additional assumptions or hypotheses are presupposed. Hypotheses should therefore not be understood in isolation, but within a broader theoretical framework.

In the social sciences, attention can be turned to public administration, where a hypothesis is formulated that bureaucrats act budget-maximising — trying to attract larger amounts to their organisation. This hypothesis can be formulated based on public choice theory, which assumes rational bureaucrats trying to maximise self-interest. Testing this hypothesis proves to be quite challenging, as measuring budget maximisation could involve surveys, observations, and much more. Budgets or documents from a given organisation could also be examined. But the question is, do these factors test the hypothesis accurately enough? In such a study, from both a methodological perspective and the Duhem-Quine thesis, it is seen that there are numerous assumptions and hypotheses associated with the formulated hypothesis; that bureaucrats act budget-maximising. When testing a hypothesis, therefore, not only the hypothesis itself is tested, but a collection of assumptions and hypotheses. In other words, it can be difficult to know exactly which idea, hypothesis, or assumption produces a particular result.

On the other hand, within the context of public administration research, a pragmatic approach would suggest considering the specific problems and challenges faced by bureaucrats in their day-to-day activities. Contrary to logical positivism or critical rationalism, the concept of truth according to pragmatism is grounded in the idea that a statement or assertion should contribute to solving problems. It is important that statements or ideas are “useful” rather than thinking of a statement as corresponding “one-to-one” with what appears in the world. If an idea, for example, does not solve the problems, it must therefore be challenged and spur new pragmatic inquiries. A statement is thus deemed true if it is useful in the long run. The truth value of a statement is thus assessed according to its pragmatic utility. However, it can be problematic for the concept of truth in pragmatism to determine when something is useful, and for whom? Pragmatism is rooted in the idea that knowledge production should be tied to solving human problems in the world. There is scepticism towards positivism’s ontology. According to pragmatism, the world is under constant development, and therefore new forms of experimentation and knowledge must arise. Pragmatic knowledge and truth are therefore understood through specific events and situations.

Thinkers within pragmatism include figures such as John Dewey, Charles Sander Pierce, and William James. Despite their philosophical differences, events are here understood here as temporally fixed and emphasise the uniqueness of various phenomena in the world, whereas situations emerge as those encountered by humans in the world in a given context, thereby determining the understanding of the world. The ontological basis for pragmatism is non-reductive or non-scientistic naturalism. Therefore, the natural sciences do not exclusively define nature; instead, everything is part of the world and nature on equal footing. At a sufficiently abstract level, it can be argued that unity of science is supported in pragmatism. Epistemologically, the world must be understood through a particular problematic situation, which, through pragmatic investigative processes, leads to understanding and solving a given problem. Science is therefore driven by the purpose of solving specific problems in the world. If the knowledge in question does not solve the problems, it must therefore be challenged and inspire new pragmatic investigations. Pragmatic investigations can also involve ensuring specific values. For example, John Dewey examined how democracy could be practised in education via his famous work, “Democracy and Education.” The pragmatic theory of truth states that a statement is true if it is useful in the long run. The truth value of a statement is assessed according to its pragmatic utility.

When discussing scientific theory and truth, we encounter various perspectives that influence our understanding of the world. These perspectives span from problem-solving in a given context (pragmatism) to investigations of truth through falsification (critical rationalism) and confirmation (logical positivism), as well as how we perceive the social world of money, law, and institutions (constructivism). In light of the developments in science, Thomas Kuhn undertook an approach to the theory of science from a more historical perspective. Kuhn, who questioned the rationality and cumulative nature of science, presents in his book, “The Structure of Scientific Revolutions,” a central thesis that science consists of periods of stable growth characterised by revisionary revolutions. Kuhn, who approached the theory of science from a more historical perspective, studied scientific development, from which he developed his understanding of science as non-rationally conducted, unlike thinkers within critical rationalism and logical positivism, who assumed that development was logical, as a result of verification (logical positivism) or falsification (critical rationalism). Kuhn argues that for scientific paradigms reflecting a given theoretical framework, there are theoretical and empirical anomalies that gradually make it difficult to maintain a particular paradigm. These paradigm shifts occur relatively rarely according to Kuhn, but when they do, they are of a revolutionary nature. Kuhn argues that scientific development does not appear to be rational or logical, but instead, scientific paradigms enjoy relative stability, followed by revisionary revolutions. However, it is often seen according to Kuhn that researchers try to explain anomalies within their current theoretical framework.

In classical economics, utility theory claims that the perceived value of gains and losses is equivalent. However, with the paradigm shift to behavioural economics (including prospect theory as developed by Daniel Kahneman and Amos Tversky), it turns out that gains and losses are not equivalent. The concept of “loss aversion” can be introduced, where the perceived value of loss is higher compared to gain. For example, a loss of $100 is perceived as a greater loss than the gain of $100. The behavioural economic paradigm here breaks with the classical economics’ assumptions that losses and gains are equivalent because market actors perceive a greater loss in utility from losses than utility gain from gains, even though it is the same amount.

When discussing science, considerations about values ​​in science must also be taken into account. Epistemic values ​​deal with knowledge and truth, including how we acquire, evaluate, and apply knowledge. Epistemic values ​​include values ​​such as reliability and truth. Non-epistemic values ​​can involve ethical, aesthetic, or emotional dimensions, concerning everything from justice to beauty. Distinguishing between these two forms helps articulate which values ​​relate to knowledge, knowledge acquisition, and cognition (epistemic values), and values ​​that relate to human morality and worldview (non-epistemic values). An example of an epistemic value ​​could be that a researcher has a dataset that highlights a trend among a defined population group, whereas a non-epistemic value ​​could be the researcher’s moral considerations about the political influence of publishing this research. Some scholars argue that researchers should incorporate non-epistemic considerations into the publication of their research, while others believe that researchers should only consider epistemic values, that is, what the scientific study actually shows.

According to Kasper Lippert-Rasmussen, if value judgments can only be justified by another value judgement, the latter value judgement must also be justified. In relation to scientific value relativism, it must be emphasised that there is a value judgement that can only be justified with reference to another value judgement. In such cases, there are two options; either the second value judgement can be justified by the first, or a third value judgement must be presented. Both of these situations can cause problems. Since the first scenario constitutes a circular argument, while the second option leads to infinite regress. In a nutshell, circular reasoning occurs when what needs to be proven is assumed. The conclusion thus follows a circular process. An example of circular reasoning could be stated as (A) “The Bible is true because it follows God’s word” and (B) “God’s word is true; it is written in the Bible.” The logical fallacy points out that A is true because B is true, and B is true because A is true, and thus, circular logic ensues. Infinite regress, on the other hand, explains an infinite process whereby the first value judgement is justified by another value judgement, and then a third, fourth, and so on infinitely. This leads to infinite regress, as the value judgments are justified by reference to additional value judgments, continuing infinitely. It is a justification chain without any actual endpoint.

Another prominent figure, who considered the role of values in science, Max Weber, pointed out that values ​​can be relevant in the social sciences in the sense that values ​​should be made objects of investigation. The second meaning of values ​​according to Weber is the so-called value relevance, that values ​​have an impact on what research communities choose to research in the first place. The third way in which values ​​become relevant in the social sciences, according to Weber, is related to goals and means, for example, how to best achieve a given goal. As the fourth way, values ​​become relevant by incorporating meaning, whereby a systematic analysis of the consequences of central values ​​and concepts can be conducted. According to Weber, meaning analysis can also lead to criticism, which is the fifth category describing how values ​​appear to be relevant in the social sciences. Weber writes about how values ​​in Protestantism, in interaction with capitalism, lead to the spread of rationalisation throughout society. Rationalisation involves a historical movement towards a world that can be more effectively controlled through calculation. It describes a process by which traditional values, emotions, and motivations are replaced by rationality and instrumental reason. Society is thus moving towards a higher degree of scientific thinking, calculation, and instrumentalisation. To summarise. First, values ​​can be relevant in the social sciences in the sense that values ​​should be made objects of investigation. Here, it can be examined how values ​​affect actors in a given field of study. Secondly, values ​​may have an impact on what research communities choose to research in the first place. Finally, values ​​become relevant by incorporating meaning, whereby a systematic analysis of the consequences of central values ​​and concepts can be conducted. Meaning analysis can also lead to criticism, which is the fifth category describing how values ​​appear to be relevant in the social sciences.

Elizabeth Anderson does not completely oppose the influence of values but instead asks whether these values can be positively directed towards epistemic values such as knowledge and truth. Thus, values can play a constructive role in understanding the world. According to Anderson, moral and political values ​​are equally distant from the truth, as some values ​​can be beneficial in guiding an investigation. However, she points out that a problem arises when values ​​are held dogmatically, leaving no room for open consideration or the ability to revise previous beliefs. For example, if a study were to be conducted on divorce, the traditional assumption that divorce is bad for all parties involved can be contrasted with the feminist assumption that based on emotional experiences, divorces can in some cases be experienced as liberating for women and children (or men). Anderson’s point is that reliance on value judgments can end up functioning as assumptions or influencing a proposed hypothesis. Anderson thus points out that bias and values ​​do not necessarily oppose understanding the world; however, a dogmatic attitude is problematic. Anderson’s argument can certainly be criticised from various epistemological perspectives, one of the most significant criticisms being that non-epistemic values ​​should be excluded from influencing scientific inquiry. Heather Douglas argues that researchers should consider the societal consequences of their research and consider non-epistemic values ​​in their decision-making process. This is because, according to Douglas, researchers have a moral responsibility for the potential effects their research findings may have on the surrounding society. To elaborate on this viewpoint, Douglas introduces several concepts related to objectivity and research ethics. Here, she writes about aspects to consider such as convergent objectivity, which means that research from different sources and perspectives should converge towards similar results to achieve credibility and reliability. Research should, according to Douglas, thus remain objective, even though it does not exclude the researcher’s awareness of potential consequences.

In light of these considerations on objectivity and research ethics, it is important to delve into the philosophical underpinnings of knowledge itself. One such perspective is offered by Plato, whose definition of knowledge states that a subject (S) knows that a proposition (P) is true if and only if: 1) S believes P, 2) S is justified in believing that P is true, and 3) P is true. Knowledge can be considered as a person having knowledge of a proposition (P) being true based on three conditions. Firstly, the person must be convinced of P. Secondly, it is necessary that the person is justified in their belief, and finally, P must be true. However, this definition can be criticised, among other things, based on the Gettier problem, which necessitates an additional condition: 4) that point 2 is not fulfilled accidentally. In connection with the last condition, there must be a clarification of what it means for something to appear as justified accidentally. The discussion of what it means to decide if something appears as accidental is ongoing in the academic community. But the Gettier problem shows overall that the three conditions, as described initially, do not constitute a sufficient basis for knowledge. Further, converging between analytical and synthetic knowledge, analytical knowledge is derived from logic and definitions, so analytical propositions are true by virtue of their meaning, for example, “all bachelors are unmarried men.” Synthetic propositions are true because they can be verified, for example, “Jens is a bachelor.” Synthetic knowledge is thus based on the collection of true propositions. Analytical truths are based on logic or their meaning. An example of an analytical truth is that “all bachelors are unmarried.” Synthetic truths are not only determined by internal truths but also by the nature of the world. An example of this is “sea water is salty.” Synthetic truths can also include political statements such as the democratic peace thesis, which states that “democracies do not go to war with each other.”

One can also distinguish between epistemic and pragmatic reasons. The former, epistemic reasons, can be described as the reasons that appear based on having evidence to believe that a proposition (P) is true. When discussing the existence of God, one could argue that there can be a pragmatic reason to believe in God, but it does not necessarily imply that there is evidence for believing that P is true — epistemic reasons. There are also many ways in which truth can be conceived and defined. For example, correspondence theory indicates that truth is grounded in our statements and science reflecting how the world is. The proposition (P) must therefore correspond to reality. What the theory says is that P must correspond to things as they appear in reality — independent of the statement made. For example, there can be correspondence between saying “the earth is round” and the actual shape of the earth. This theory also fits well with fact objectivism. A critique of the theory can arise by questioning how it is judged that a statement corresponds to reality. As an attempt to solve this problem, Ludwig Wittgenstein can be highlighted, who argued that if a picture is taken out in reality, there is a reality relationship — or likeness — between reality and the picture. Similarly, when talking about sentences, there can be a likeness between sentences and reality, e.g., “the cat is lying on the mat.” However, Wittgenstein later ended up rejecting this theory. Coherence theory indicates that the criterion for truth becomes that the statement must align with other beliefs about the world. Here, the aim is towards the greatest possible coherence or logical agreement. A statement must be judged as to whether it is true in relation to other statements. For example, Jupiter’s movement can be explained in the context of a statement about gravity. A statement consists of coherence relations, including a statement being coherent only if they are consistent with each other — but sentences can also be consistent without an inherent connection. Therefore, more than just consistency between two statements is needed. An example of coherence could be: Socrates is mortal, all humans are mortal. Socrates is human.

Consensus theory describes how a statement is true if consensus can be reached about the statement in question. A statement is considered true here if consensus can be reached about the statement. The criticism here can be, who should there be consensus among? For example, when looking at COVID vaccines, it can be asked whether there should be consensus among medical professionals or consensus among anti-vaccine groups and medical professionals. What does it mean for a statement to be true for group 1 but false for group 2? And how do we deal with this problem? This theory also fits as a perspective on fact constructivism. Pragmatic truth theory, on the other hand, describes how a statement is only true if it appears useful from a long-term perspective. A statement is true if it is useful in the long run. The truth value of a statement is assessed according to its pragmatic utility. The pragmatic truth theory can thus be understood as an approach to the world where knowledge and understanding that are useful are sought — so what is the pragmatic value? A criticism of this could be to ask when something is useful and for whom it is useful? For example, a theory could be useful for some people but not useful for others.

Considering the minimalist truth theory, it describes how a statement (P) is true if and only if P. This theory of truth therefore does not focus on truth in terms of coherence or utility. The theory is minimalist, arguing that “p holds: ‘p’ is true if and only if p.” Moving on, the epistemology of standpoint theory describes how the knowledge that can be acquired is linked to non-epistemic factors. Marxists argue that the non-epistemic factors that apply depend on whether one belongs to the working class or the bourgeoisie. Briana Toole argues that non-epistemic factors are relevant in the context of examining social identities, whether a given person belongs to an oppressive or oppressed group. Here, Toole illuminates two factors: how knowledge is situated and the examination of epistemic oppression or exploitation through the situated knowledge thesis. The situated knowledge thesis states: “there exist statements, p, for which whether an epistemic agent is in a position to know p depends on non-epistemic social facts about the agent.” Toole argues that there are non-epistemic facts, for example, whether a given person belongs to a minority or majority, here referred to as social identity. For example, there may be a discussion about whether a particular statement is an expression of “gaslighting.” For instance, if one is heterosexual and tells jokes about homosexuals, it can be questioned whether this is an expression of “gaslighting” or not, according to Toole’s argument. According to the situated knowledge thesis, it can be argued that there is a hermeneutic injustice or asymmetry, where the dominant group also possesses a privileged epistemic situation according to Toole. Epistemic oppression occurs, for example, when women are excluded from the research world or queer individuals are not allowed to enter certain workplaces. Epistemic oppression can be described as a persistent epistemic exclusion of a particular group, which often finds itself in a vulnerable situation or minority position. Thus, power balance may appear disproportionate.

In science, there are also ways in which the state of affairs is explained. The explanandum includes what is sought to be explained, while the explanans is what explains. Explanandum revolves around what is sought to be explained — i.e., what may be relevant to investigate. Explanans explains the explanandum. Token explanandum involves a singular point or event. Here, there is a singular explanation. An example of token explanandum in political science may relate to questions about why women voted more left-wing than men in the Swedish election. Type explanandum involves general trends and explanations. Here, an explanation for a general trend can be investigated; why women are increasingly voting left-wing in the recent elections in Sweden. A bad explanation relies on false assumptions, explanans does not explain explanandum, and explanans does not provide knowledge that calls for explanation. In connection with causal explanations, one can discuss how Hume highlights the problem of induction. According to Hume, a person can observe a series of events, but whatever the causality is, it is not directly experienced. Therefore, Hume’s philosophy calls for questioning whether causality can be included in explanations. Approaches popular in the scientific approach may involve causal inference, where causal dependencies are followed through statistical examination of the world. Here, inferences can be made from data about populations. Additionally, randomised control trials can be conducted, where a control and treatment group are established to examine the causality of the treatment. The process of randomisation and comparison between the control and treatment groups helps to exclude spurious relationships. Carl Hempel listed three components involved in a causal explanation: 1) triggering cause, 2) a series of initial conditions, and 3) one or more laws. This explanatory model is based on the regularity theory of causes.

There may also be questions about regularities in science. Laws can be described as regularities that occur. This can be contrasted with purely coincidental occurrences. In social science, there may be questions about whether there are laws in social science or “law-like regularities.” Unlike natural laws such as gravity, it can be argued that while natural science can predict 100 out of 100 times that an apple will fall from the tree, social science cannot necessarily identify such strong regularities. The laws in social science are therefore not like those in the world of physics or chemistry; they are not universal. In social science, “regularities” appear more in a specific context, where temporality, culture, system, etc., can be considered. According to the deductive-nomological model, everything still appears as inductive. Finding evidence for hypotheses would require inductive forms of argumentation. Many social science researchers do not give up on exploring causal regularities. This includes looking at types of explanations, statistical and quantitative methods, and randomised experiments. Process tracing and the construction of narratives are an example of token explanation — combining evidence and exploring causal mechanisms.

In July 1898, France and Britain were on the brink of war over territory, but surprisingly, the expected war did not break out. Sharon Crasnow presents three lawful hypotheses for the “Fashoda incident”: 1) neorealism explanation, asymmetric power balance, 2) democratic peace theory, and 3) informational asymmetry. Crasnow writes about narrative as explanation, arguing according to process tracing that narrative constitutes the cognitive work of making an explanation credible. Here, consideration of alternative hypotheses is facilitated, and the relationship between the given evidence and explanation is classified. Another form of explanation is purpose explanation, which involves referring to internal effects. Humans act as rational agents seeking to maximise self-interest. This means that according to this model of understanding, people act based on a purposeful rationale. There are two different kinds of functional explanations. One described by Lamarc is the direct function, while the other via Charles Darwin is based on the indirect function. For example, if one looks at the long neck of the giraffe, the direct functional explanation would be that this trait is functional in eating leaves from the trees. Darwin, on the other hand, has an indirect functional explanation, which describes how the giraffe has evolved through evolution, as it leads to survival and reproduction. Functional explanations could be in political science that nationalism is good for cohesion among people. Jon Elster examines when an acceptable functional explanation emerges. In this context, Elster presents a framework:

• Y is an effect of X
• Y is beneficial for Z
• Y is unintended by the actors producing X
• Y (or at least the causal relationship between X and Y) is unknown to the actors in Z
• Y maintains X through a causal feedback mechanism going through Z

In addition to different types of explanations, social science also grapples with the dichotomy between methodological individualism and holism. Methodological individualism rejects so-called superindividual social facts. According to this approach, social phenomena can be reduced to aspects of individual interaction. This approach thus focuses on observable individuals to examine the social world in accordance with the requirements of empiricism. In this way, a foundation for observable investigations is established. Methodological individualism can also be described as being linked to unity science and the possibility of reducing social phenomena to individual psychological, biological, chemical, and physical phenomena. Methodological individualism is found in economics. For example, Adam Smith argued that individual economic actors engage in a market of rational individual production and consumption behaviour. What motivates the cobbler, the baker, and others are the conditions for sale and purchase in the market according to Smith. A concept formulated by Smith, “the invisible hand,” describes through a metaphor how the market is self-regulating according to a spontaneous order. Here, one can speak of individuals maximising self-interest, and market equilibrium based on supply and demand for prices, labour, and more.

When discussing behaviourism or evolutionary psychology — which are distinct directions — it can be argued that these fields ontologically are grounded in a naturalistic and methodological individualistic understanding of the world. Epistemologically, there is a focus on an empirical understanding of individual behaviour. There is a pragmatic value in being able to reduce everything to the individual level. The reduction thesis is central to methodological individualism. According to methodological individualism, social entities do not have agency. It is only individuals who are treated as agents. The principle of exhaustiveness describes how individuals and their relationships are exhaustive in order to investigate social phenomena. Thus, the social world consists of individuals according to this principle. The principle of determination describes how individuals determine the social world through actions. Reference to individuals constitutes a condition for an explanation and examination of social phenomena.

On the other hand, methodological holism opposes “individualistic reduction”; in fact, science should be “non-reductive.” Methodological holists will also point out that unlike natural science, which has natural laws, social science does not have these same kinds of laws, and therefore, social science must more closely examine the world holistically. The core assertion of methodological holism is therefore that even though social phenomena may depend on individuals, social science cannot be reduced to individual behaviour. This individualistic understanding of the world is therefore not sufficient to investigate social facts. When natural and social sciences are compared, it can be argued, according to this approach, that methodological holism is something that creates a distinction between social science and the natural sciences. Social facts are understood as social non-reducible phenomena. It can be said that there is more focus on the whole than on the parts or individuals themselves. For example, Émile Durkheim found that even though suicide is an individual act, there is a social influence or explanation that affects the individual’s decision about suicide. Therefore, it can be argued that methodological holism can more effectively accommodate social reality.

In social science, the multiple realisability argument states that there are various ways in which a social phenomenon can be realised. There are thus countless ways in which a given social phenomenon can occur. Phenomena can thus arise from many different combinations. But why is multiple realisability relevant? First and foremost, the argument shows that there is no specific constellation of individual phenomena that explains a social fact. The supervening phenomenon can be realised in several different ways. This blocks the reduction of individual and social phenomena. Multiple realisability thus challenges parts of methodological individualism. For example, the reduction to individual actions in a given field of study is opposed. Arsenal, the football club, for instance, can according to methodological individualism be reduced to individuals, e.g., football players. But here, there is a risk of oversimplifying the football club, as it could be expressed through many different constellations. In principle, all the “insides” of Arsenal could be replaced, and it would still be the football club, Arsenal.

Social sciences tend to, from an individualistic standpoint, think of people as rational actors. This is seen, for example, in rational choice theory or classical economic theory. However, evolutionary psychology, for example, does not subscribe to the idea that humans are rational actors. Moreover, there is not always a sharp distinction between individualism and holism in investigating a phenomenon. For example, an educational or career choice can be reduced to an individualistic phenomenon, but at the same time, from a holistic standpoint, there are norms, culture, and ideas about what constitutes a good career. Sometimes the distinction between individualism and holism can probably be partially graded. Considering a methodologically individualist approach, we can introduce B.F. Skinner as an example; a behaviourist known for his laboratory experiments where he stimulated animals (e.g., rats, monkeys, and pigeons) to exhibit complex behaviour. These studies represent a key concept in behaviourist thinking about human behaviour as conditioned by stimuli and respective positive and negative responses from external environments.

Ontologically, behaviourism starts from a naturalistic and methodological individualistic perspective, whereby individual biological organisms emerge as the primary focus. The social world is thus part of the physical world and therefore also causally determined. Epistemologically, the social world is thus recognised through the observation of individual actors’ behaviour. Skinner’s behaviourism arose within neuroscience. Considering Ivan Pavlov’s famous experiment with dogs (“Pavlov’s dogs”), it involves a systematic investigation where a bell rang every time the dogs in the experiment were fed. In this way, the dogs began to associate the sound of the bell with the presence of food, and thus, they salivated without seeing or smelling the food when the bell rang. Pavlov’s classical conditioning describes a type of learning whereby a conditioned stimulus (e.g., the sound of the bell) can lead to a conditioned response (e.g., production of saliva). This requires, as described in Pavlov’s study, that the dogs must first learn to associate the food with the sound of the bell. Radical behaviourism requires that there is behavioural evidence for psychological hypotheses. Centrally, there is the assumption that positive reinforcement (e.g., pleasure) leads to repetition, while negative responses (e.g., pain) lead to avoidance.

The Law of Effect describes that behaviour is determined by external stimuli and hence positive and negative responses to behaviour. If a behaviour is rewarded, this implies that the behaviour will be repeated with increased frequency. On the contrary, if the behaviour is punished, the frequency decreases. The Law of Effect describes a fundamental law in behaviourism with its mechanical worldview. By comparison, the Law of Folk Psychology (L) can be formulated, which deals with how a person acts based on a desired outcome. The explanation goes: If a person (x) desires a specific result (d), then the person will take action (a) to achieve d, provided that the person is convinced that a is a means to achieve d. This can be described as a form of rationality, where people act based on a given purpose. Folk psychology is based on the idea that behaviour can be explained by concepts such as beliefs, desires, intentions, and other mental states. The assumption here is that people act based on specific intentions. Epistemically, it is easier to understand behaviour than the underlying processes that occur in the mind. It is often easier to empirically measure behaviour than underlying mental states according to behaviourists. To provide causal explanations, the cause must precede the effect in behaviourism. Behaviorism explains the cause, referring to the effect (e.g., external stimuli such as pleasure or pain). Behaviorism can primarily be explained as a general behavioural psychological approach to investigating the world. However, there is also relevance here within the field of political science. According to behaviourism, the intent is to investigate political behaviour, focusing on human behaviour rather than what they think. In this context, David Easton formulated behavioural “creeds” in the 1960s, which state:

• Regularities: Observable uniformity appears in political behaviour, which can be expressed in theories or generalisations.
• Verification: The generalisations in question can be tested, focusing on a given behaviour.
• Techniques: There must be methods for obtaining and interpreting empirical data.
• Quantification: It must be possible to measure and quantify data.
• Values: A distinction must be made between ethical evaluation and empirical explanation, as they involve different approaches or claims to analysing the world.
• Systematisation: There must be a focus on creating well-structured systematic knowledge about theory and research.
• Pure science: Research in political behaviour logically precedes and forms the basis for investigating and solving political problems.
• Integration: The social sciences must have the capacity to integrate results from other disciplines.

Behaviorism thus constitutes examples of a naturalistic mechanical worldview, whereby laws for human behaviour can be established, including within the field of political science. Here, the focus is on the ability to predict behaviour, incorporating regularities and the like. Human beings are understood as entities that are externally determined and influenced by various factors in the external environment.

Considering evolutionary psychology, it is a part of psychology that can be described as naturalistic and methodologically individualistic. Here, behaviour is described with reference to the human “state of nature,” i.e., the environment to which humans have evolutionarily adapted. The behaviour exhibited by humans is thus governed by the evolutionary context, how humans have adapted according to their ancestors’ living conditions when they lived as tribal people. Many of the characteristics humans have developed long before developments such as industrialised society. Michael Bang Petersen, Daniel Sznycer, Aaron Sell, Leda Cosmides, and John Tooby present some evolutionary psychological arguments in their research article “The Ancestral Logic of Politics: Upper-Body Strength Regulates Men’s Assertion of Self-Interest Over Economic Redistribution.” For example, a hypothesis is put forward that high upper-body strength can improve men’s bargaining power over scarce resources. This leads to a prediction that men with higher upper-body strength will hold powerful political positions, whereas men with lower strength will hold less powerful positions. The latter prediction is that an advantage in physical strength for women does not correlate with their position. In modern society, however, many other factors appear besides physical advantages. Physical strength does not correlate perfectly with power, as there are many men with low socioeconomic status but high physical strength, and men with high socioeconomic status who do not have a strong physical appearance. Despite this, it was found that among men with lower socioeconomic status, strength predicted increased support for redistribution. However, among men with higher socioeconomic status, strength predicted increased resistance to redistribution. Cross-cultural studies are also particularly interesting as they can help eliminate the idea that cultures in a given population lead to a given outcome. If a specific effect is seen across cultures, there is evidence according to cross-cultural studies for a given outcome.

Transitioning from the evolutionary psychology perspective to phenomenology, we move from a naturalistic, individualistic approach rooted in the evolutionary context to a philosophy of science of human consciousness and experience. Taking on a less mechanical worldview, critiques of ideas that science presents dualisms such as mind and body, and subject and object, are found within phenomenology. Phenomenology also does not seek to distinguish between the so-called “thing-in-itself” and the “thing-for-us.” Additionally, phenomenology criticises science’s observer perspective as third-person and instead focuses on the first-person perspective, how the “self” or “I” orients itself towards and experiences the world. How individuals ascribe meaning to their surroundings and actions constitutes an interest for the researcher. The central focus of phenomenology is to seek understanding of how phenomena appear in human interaction with the world. Thus, there is a focus on human consciousness structures. Phenomenology aims to delineate the immediate meaning-making that occurs as well as experience. Ontologically, phenomenology assumes that the subject is situated in the world, and epistemologically, the world is understood via the first-person perspective. Edmund Husserl formulates the transcendental ‘ego,’ which is extended in space and time and enables various experiences of consciousness. Martin Heidegger writes about Dasein (being-there), which is described as a precondition for creating meaning in the presence in the world. Here, Heidegger precedes what can be thought of as the traditional distinction between subject and object. Jean-Paul Sartre is concerned with the ability to create meaning in a world that seems meaningless or ambiguous. Here, Sartre’s argument that existence precedes essence can be highlighted. On the other hand, Maurice Merleau-Ponty deals with how bodily experiences constitute an essential part of cognition. Therefore, consciousness cannot be thought of as separate from the body, as the body is part of it.

Lifeworld describes the everyday world that most people take for granted or do not typically question. The practices of a social, practical, and bodily nature that constitute part of everyday life in an ordinary human life emphasise the concept of lifeworld. The concept of lifeworld also helps to underscore that the world becomes the object of consciousness’ directedness and how individuals experience and attribute meaning in their everyday ordinary presences. These meaning-making practices are necessary in conducting science, politics, and more, as not only goals and means, quantification, measurement, or the like are considered. Jürgen Habermas formulates, among other things, a well-known critique of the colonisation of the lifeworld, where the system world (state and market) gradually colonises the lifeworld (here referring to critical theory as well). The method of phenomenology is thus grounded in examining phenomena as they appear in the subject’s attitudes, ideas, experiences, and cognitions. The relationship between phenomenon and consciousness constitutes the focal point here. Epoché describes the natural or basic assumption of the world that is paused in order to see the world with fresh eyes. In phenomenological research, this is known as the process where assumptions are suspended in order to explain a phenomenon in terms of its inherent system of meaning. In contrast, ethnomethodology uses ethnography as an approach to get behind what is often taken for granted in cultural contexts. Thick descriptions are made to illuminate the nuances in the field of study, while methods such as fieldwork, participant observation, and lifeworld interviews can be used.

Where hermeneutics focuses on interpretation and the transmitted intersubjective language as well as culture, history, and traditions, phenomenology focuses more on subjectivity and consciousness, how phenomena are experienced. Phenomenology is more directed towards understanding human consciousness as phenomena appear to them, whereas hermeneutics places significant emphasis on how things can be interpreted in relation to language, culture, and history. From an etymological perspective, the word hermeneutics originates from the Greek word hermēneuein, meaning to translate or interpret. The hermeneutic approach was originally used in the interpretation of religious and legal texts. However, Hans-Georg Gadamer in his Truth and Method developed central concepts for philosophical hermeneutics, which are used in the social sciences today. The hermeneutic circle emerges as a central methodological approach within hermeneutics. The hermeneutic circle highlights that the interpretation of a given phenomenon occurs in a movement between part and whole. Gadamer considered humans as fundamentally interpretive beings. That is, the active relationship of humans to the world is interpretive. Ontologically, humans are thus assumed to be interpretive and meaning-making beings, where epistemologically, knowledge can be acquired through understanding and interpretation. From a methodological standpoint, this entails interpreting texts or social phenomena in order to arrive at understanding and knowledge from existing preconceptions.

The hermeneutic circle describes a movement between part and whole. This is where horizons of meaning can be identified and their possible merging. In such a hermeneutic investigation, particular areas such as language, history, concepts, norms, worldviews, and more can be focused on. The hermeneutic circle refers to the idea that a text or a social phenomenon can be interpreted in circular movements, whereby the part assumes a horizon of understanding for the whole, and the whole for the part. Through this approach, the hermeneutic process can serve as a basis for interpreting complex social or political questions. In hermeneutics, the encounter with the foreign horizon is articulated in relation to the interpreter’s horizon, as a merging of horizons. For example, a historical phenomenon or an unknown culture — a foreign horizon — can be related to the interpreter’s horizon of pre-understandings. The phenomenon thus enters the interpreter’s horizon, while the interpreter enters the phenomenon’s horizon, whereby the respective horizons “merge.”

Hermeneutics is not only involved in the interpretation of legal or religious texts. Hermeneutics is also more than just an approach that can be used within philosophy. In political science, it may be relevant to have an interpretive approach to a given field of study. For example, hermeneutics can contribute to interpreting power — “the hermeneutics of suspicion” — where the hidden or unsaid is perceived as significant. Considerations about gender, contracts, capitalism, and much more can be involved here. Hermeneutics is not about empathy closely through individuals’ consciousness structures; instead, hermeneutics revolves around how social phenomena can be interpreted in the study of the language, concepts, and understandings various actors have about themselves and the world. This also includes the study of self- and world-understanding. In connection with hermeneutics, it is also essential according to Michael T. Gibbons to understand how language and concepts can differ across temporal and spatial dimensions. Gibbons argues that the political science approach has failed in the sense that it has not adopted a significant focus on developing an interpretive approach to political research. Gibbons believes that the understanding of political life through an interpretive approach has stagnated and not progressed. Thus, hermeneutics offers an interpretive method or at least a complementary approach to investigating political phenomena. Gibbons highlights criticisms of behaviourism and Western rationalist assumptions since the interpretation of language and concepts constitutes a significant area in political studies, historical-linguistic influences must be considered.

Another critical approach to examining society can be found in the origins of The Institute for Social Research, Frankfurt am Main, also known as the Frankfurt School, founded in the 1920s. Critical theory is concerned, among other things, with criticising the capitalist society’s mode of production and dominance and how the ruling class exploits the working class, which lacks class consciousness. The purpose of critical theory is ultimately emancipation or liberation from the status quo, especially the capitalist mode of production and alienation. One of the most famous works among many within the Frankfurt School is “Dialectic of Enlightenment” (Dialektik der Aufklärung) by Max Horkheimer and Theodor Adorno. In this book, Adorno and Horkheimer criticise the Enlightenment project and instrumental reason, which, through World War II, brought chaos and destruction instead of liberation, which was the goal of the Enlightenment project. In this work, Adorno and Horkheimer also criticise a disenchantment of the world as a result of the instrumentation and objectification of the world. This pessimistic social diagnosis should especially be seen in light of the ruins and destruction of World War II, and thus, the historical and cultural context is relevant to this interpretation. Adorno and Horkheimer also criticise the culture industry, whereby film, music, and the like are produced through an industrial logic (“film industry” and “music industry”). Through this industrial logic, the production of cultural products becomes mass-produced with profit as the primary aim. Here again, criticism is directed at a means-ends rationality and the capitalist mode of production.

Habermas, in the later generation of the Frankfurt School, can be said to develop Immanuel Kant’s deontological ethics, which, in simple terms, is based on the maxim that one must act in such a way that one’s actions can become a universal law. Instead of completely rejecting the project of enlightenment, Habermas moves in his discourse ethics from an instrumental rationality to a communicative rationality. Thus, Habermas’ ethics argues that in order to establish a norm, participants affected by it must engage in a domination-free conversation, allowing the best argument to arrive at a valid norm. This places a focus on rationality among members of society rather than rationality located in the individual. Habermas’ rationality thus appears as communicative action, where free conversation is about mutual understanding and valid arguments. It can be argued that the intersubjective occurs in this domination-free conversation. Communicative action must be factually true, normatively correct, and/or express the speaker’s actual convictions. Through this domination-free conversation, the actor must engage with their own and others’ understandings based on valid claims. This discursive practice can be said to contain an emancipatory logic, where there is a continuous revision of traditions, norms, etc. Habermas also formulates a well-known critique of the colonisation of the lifeworld, where the system world (state and market) gradually colonises the lifeworld. The system world operates via codes or logics, for example, through media such as money (the market) and power (the state). According to Habermas, the colonisation of the lifeworld can lead to a loss of meaning. According to Habermas, the system world, including the economic system, political system, etc., can colonise the lifeworld, leading to a loss of autonomy within the lifeworld. The colonisation of the lifeworld displaces the lifeworld. Decoupling between system and lifeworld describes the tendency where lifeworld and system world move further apart from each other with their distinct logics, goals, and rationality. Decoupling thus does not involve a liberation from the system world, but instead the “understanding” or “dialogue” between the two spheres results in the system world not taking into account lifeworld perspectives. Speaking of critique, external criticism involves criticism of external norms, whereas internal criticism describes internal norms, including the non-fulfillment of these norms. Immanent criticism is thus an internal critique in interaction with an investigation of why certain norms or principles are not fulfilled. Immanent criticism is therefore about examining what actually appears as an obstacle to realising certain ideals or principles (e.g., freedom). Through this critical approach, a view can be directed towards cultural conditions such as ideological production, or structural conditions, for example, the society’s economic system, including vested interests, modes of production, etc. Moreover, there can be talks about personal limitations regarding how certain individuals show a lack of self-insight or awareness according to some theorists.

Considering the structuralist perspective, which focuses on the systematic analysis of language and society, we transition to a new form of societal analysis, and philosophy of science. Structuralism is based on a language theory developed by Ferdinand de Saussure. According to Saussure, language constitutes a closed structure of meaning. It is the structure of language that is highlighted as the scientific object. The meaning of a word does not come solely from what it represents but from its relational place in a structure of words. For example, a word like “bread” is not understood in isolation but in relation to other words such as “oven,” “baker,” “cakes,” and “food.” Structuralism in social science is a theory about whether there are structures of meanings in society that are not available to people but still ‘force’ people to interact and relate to the environment in a specific way. Structuralism argues that people are subject to structural conditions, and structures are the focal point rather than individual actors. In social science, this means that instead of looking at the sum of individual actions, structuralism points towards making structural analyses of society’s systems or structures. This could involve studies related to institutionalisation or the state apparatus. As an example of structuralism in social science, structuralist Marxists argue that the base, including the economy, determines the superstructure, the state apparatus, and the ideology of the status quo. Louis Althusser is an example of a thinker who was a structuralist Marxist. By focusing on language and the world, it can be argued that constructivist elements emerge. However, structuralism does not peculiarly focus on how humans create and interpret the world in the same way as constructivism or hermeneutics.

Post-structuralism holds onto structuralism’s anti-individualism and still emphasises that the essential emerges in the relation between the parts. However, there is a difference in that post-structuralists reject the idea that structures or systems assume a fixed form. Instead, structures are described by post-structuralists as decentralised forms where there is neither a fixed centre, essence, or the like. A system, therefore, cannot achieve a fixed stable form as it is constantly changing in how it interacts with society, including areas such as politics and ideology. According to post-structuralism, structures are not understood as closed but as changeable and incomplete. Consequently, even the meanings of words are explained to be more open and unstable than assumed by structuralism. Meanings and identities are created through language. In scientific discourse, it can be analysed, according to thinkers such as Michel Foucault, how language and structures contribute to articulating madness, sexuality, and disease. In scientific discourses, a range of ‘subject positions’ according to Foucault, such as when a person is categorised as mad or sick, appears. Objects, such as patients and citizens, are thus offered subject positions. Biopolitics refers here to how the political and administrative evaluates life forms and their characteristics. It is about the interaction between political power and biological autonomy. Examples include public health policies, reproductive rights, and biometric data and surveillance. Genealogy describes how units, subjects, or identities have arisen and changed throughout history. For example, national states or the international community can be examined here. Essence describes the “innermost” of something, what can be described as the crucial relationship that characterises a given thing. According to Marxism, the essence of contemporary society can be described as capitalist.

Moving on to another paradigm of social analysis, feminism serves to spotlight gender disparities and oppression. Feminist empiricism includes, as the name suggests, empiricism and feminism. The former involves an empirical focus on observations and data collection, while feminism helps guide the research. Feminist empiricism is a perspective within feminist studies that combines goals and observations in feminism with scientific methods and empiricism. Hume’s body of work is related with empiricism, where empiricism forms the core of scientific practice. Feminism comes into play as values can guide empirical observations. Feminist empiricists could, for example, criticise medical experiments, which are empirically based, but only include young men as subjects of study. This is problematic from both a feminist and medical standpoint since it can lead to women not receiving treatment tailored to their physiology. Feminism can be said to attempt to salvage the objectivity of science by incorporating tools through which contextual values can be articulated or controlled.

Feminist empiricists are cautious about incorporating non-epistemic values into scientific practice. For example, if an empirical study indicates that a portion of women perceive divorce as liberating, it does not necessarily imply that this applies to all women. Epistemic values can be described as clear and simple, often referred to as objective. Non-epistemic values can address moral and political values, happiness, health, or societal risks. As an example, a researcher might consider the societal risk of publishing their data. Anderson does not completely oppose the influence of values but instead asks whether these values can be positively directed towards epistemic values such as knowledge and truth. Thus, values can play a constructive role in understanding the world. According to Anderson, moral and political values are equally far from the truth since some values can be beneficial in guiding an inquiry. However, she points out that a problem arises when values are held dogmatically, leading to a lack of open-mindedness or ability to revise previous beliefs. For example, if a study on divorce were to be conducted, the traditional assumption that divorce is detrimental to all parties involved could be contrasted with the feminist assumption that based on emotional experiences, divorces can be liberating for women and children in certain cases. Anderson’s point is that reliance on value judgments can end up functioning as assumptions or influencing a hypothesis put forth. Thus, Anderson emphasises that bias and values do not necessarily oppose understanding the world, but a dogmatic attitude is problematic. Anderson’s argument can certainly be criticised from various epistemological perspectives, one of the most significant criticisms being that non-epistemic values should be excluded from influencing scientific inquiry.

Helen Longino argues that, for example, sexist and racist ideas can negatively influence science, and therefore research must acknowledge the inevitable influence of values in science. To address this problem, Longino presents in her socialised theory of science (contextual empiricism) procedural objectivity, as there will always be values in science. Therefore, it is not a question of achieving complete value neutrality, but instead, hypotheses should be subjected to critical dialogue and assessment. For procedural objectivity to exist, the following factors must be included: publicity, openness, standard, and moderated equality. Moderated equality implies that different perspectives are heard, but it does not necessarily mean that all researchers (due to various research backgrounds) are weighted equally. One of the differences between Anderson and Longino appears regarding contextual values. Although both Longino and Anderson acknowledge the influence of values in science, Anderson argues that this can be controlled. Anderson believes that a distinction can be made between legitimate and illegitimate use of values in scientific research. Illegitimate values are understood as something that predetermined the conclusions of the research project. The problem of underdetermination describes how research is based on assumptions about the validity of previous research. When testing, it is not only about the hypothesis but also about the assumptions that form the foundation of the theoretical constructs. Therefore, it cannot be certain whether it is the individual hypothesis that is confirmed or refuted by the empirical data in question. Longino positions herself within feminist empiricism, proposing a contextual empiricism that argues that contextual values will always influence research. Contextual values describe values determined by social and cultural context. Considering Donna Haraway’s philosophy, it is argued that knowledge is influenced by cultural, social, and historical positions. There is no such thing as a “god trick,” which is why a final objective standpoint cannot be reached. Logical positivism assumes a “view from nowhere,” and social constructivism assumes a “view from everywhere.” Both of these viewpoints Haraway calls “god tricks.” The criticism of this “god trick” is that knowledge is always situated, and by this “god trick,” the researcher remains unaware of how contextual values affect research projects.

Unlike feminist empiricism, which evaluates the empirical problems of research, feminist standpoint theory begins from marginalised groups, focusing on the implications of scientific research for marginalised groups. Among the standpoint theorists are philosophers such as Sandra Harding and Alison Wylie, as well as sociologist Patricia Hill Collins. Standpoint theory acknowledges that knowledge is situated. The dominant group (the majority) does not necessarily have the experiences or abilities to understand the advantageous position they hold. By including marginalised perspectives, scientific practice can come closer to achieving a full picture of the research field. To achieve a nuanced scientific research practice, there is thus a focus on implications for marginalised or vulnerable groups regarding research practice.

Standpoint theory starts from specifically marginalised standpoints. Therefore, there should be a focus on the groups that are often overlooked or marginalised in society or science. According to standpoint theorists, all knowledge is situated. There is thus no so-called “view from nowhere” according to standpoint theorists, but there are some standpoints or “starting points” that are preferable from an epistemological standpoint. Objectivity according to standpoint theorists should not be understood in the traditional sense, but a focus on a collection of subjectivities or marginalised groups can move research practice closer towards objectivity. It is therefore essential to acknowledge a wide range of social groups and positions to understand the reality that science investigates. The idea here according to standpoint theory is to seek a comprehensive picture of a complex world, especially where marginalised groups are examined in understanding the social world. The problem is that science often reflects the standpoint of the dominant group.

As an example, Collins is known for her work “Black Feminist Thought,” which establishes a feminist theory focusing on women of different genders. Collins’ standpoint or “starting point” is Black women and their experiences. However, it also involves more than that, as it requires engaging in dialogue and reflections on these experiences, also in interaction with the majority. With reference to Wylie, three standpoint theoretical theses can be put forward: (1) situated knowledge, (2) inverted privilege, and (3) accomplishment. Situated knowledge describes how all research practices are influenced by social, cultural, and historical factors. Inverted privilege describes how individuals in marginalised groups have access to other so-called epistemological resources. This implies that they have access to evidence, knowledge, and interpretive possibilities from a different perspective than the dominant worldview. The latter accomplishment encompasses the fact that situated experience is essential in research practice, but at the same time, it also involves the development of an epistemologically relevant standpoint in analysing the organisation of society. Wylie’s standpoint starts with indigenous groups in Canada, among other places. She conducts critical practice focusing on these indigenous groups. To compare with Longino’s and Anderson’s feminist empiricism, Harding is more focused on how marginalised groups are affected, whereas Longino and Anderson focus on maintaining empirical and objective science. Harding writes about strong objectivity, stating that certain social situations are better than others in posing questions and establishing knowledge. Harding also refers to Collins’ idea of dialogue as being proactive. Strong objectivity is about recognising this situated knowledge and acknowledging the relevance of drawing on a marginalised group to gain comprehensive knowledge of the social world.

As a critique of standpoint theory, it can be questioned, for example, who is marginalised? And what about, for example, white men working in daycare or other pedagogical work? How can the researcher best rank or categorise marginalised groups? When the studies are initiated, there is a risk that the researcher’s own subjectivity will influence who is considered marginalised or not. Standpoint theorists would likely counter-argue that all knowledge is situated, and thus the question is more about engaging in critical dialogue about contextual values. Standpoint theory can also be criticised for its ethnocentric element in scientific approach. There may also be a risk of a narrow focus on selected groups depending on the context and research project. Harding would probably defend this criticism by arguing that standpoint theory is a legitimate scientific position, as objectivity is not the same as value neutrality. The goal is thus to achieve complex knowledge that embraces diversity. Strong objectivity is about recognising this situated knowledge and acknowledging the relevance of drawing on a marginalised group to gain comprehensive knowledge of the social world. Considering what may be labelled “feminist epistemology”, we examine how everyday knowledge and communication can have implications for gender understandings, for example, the perpetuation of its inequality. Epistemologically, it is assumed here that all knowledge is situated, whereby factors such as social position and power can influence who and what is considered knowledgeable and reliable. Further, intersectional feminism describes a branch of feminist theory that recognises that not only gender but also other factors such as race, class, and culture play a significant role in social positions and experiences.

Shifting from the analysis of standpoint theory and feminist epistemology, we now delve into Roy Bhaskar’s critical realism, which places considerations related to philosophy of science at the forefront. Bhaskar has developed the scientific position called critical realism. Bhaskar places ontological considerations at the centre of critical realism, as this scientific theoretical position questions the nature of reality. When talking about ontology, a distinction can be made between the realistic ontology found in naturalism and the ontological understanding according to idealism, which argues that reality is influenced by human cognition. Although Bhaskar can be said to belong to the former — the realistic ontology — this is done with various reservations. Bhaskar was critical of the realism present in positivism’s empirical realism. Hence the term critical realism. According to this critical realism, Bhaskar considered the ontology in positivism to be flat and lacking nuance, and therefore developed a so-called deep ontology consisting of multiple layers.

Bhaskar, who was critical of positivism’s realism, argued that there is a reality independent of human perceptions, which humans cannot fully access. Thus, humans only see the “tip of the iceberg” in their interaction with reality. To articulate this argument, reference can be made to the intransitive and transitive. In this context, the intransitive can be explained as a designation of science’s objects, which is the objective reality that the research process seeks to get as close to as possible. On the other hand, the transitive designates knowledge objects that can be described depending on the contexts in which they appear. These temporal and spatial contexts involve various hypotheses or theories about reality. For example, various theories can be put forth about the shape of the Earth, but these theories do not change the intransitive object; the actual shape of the Earth. This distinction between the intransitive (realistic ontology) and the transitive (relativistic epistemology) is a significant starting point for critical realism. Essentially, when talking about reality independent of human perception, Bhaskar highlights three so-called domains of reality. Here, three domains appear in what can be termed as deep reality; the empirical (what can be observed), the actual (what exists independently of observation), and the real (mechanisms that produce the actual).

Rationality is another debated area within science and philosophy. Rationality can encompass several areas from rational argumentation to rational economic management in companies. Rationality can also describe society’s increasing rationalisation or how rational thinking is applied at micro, meso, and macro levels. Thus, there are different understandings of what rationality is and where and how it appears. There are different forms of rationality. Habermas’s communicative rationality describes how participants in a domination-free conversation arrive at norms, including his discourse ethics. When talking about rationality, a distinction can be made between practical and theoretical rationality. Practical rationality describes how actions can be evaluated in terms of their practical rationality, while opinions and beliefs are assessed as theoretically rational. Practical rationality describes actual actions, whereas theoretical rationality refers to opinions. Theoretical versus practical rationality can be in opposition. For example, one’s theoretical rationality may say that one should not smoke cigarettes, while practical action leads to the person smoking anyway, thus resulting in practical irrationality. So, there may be situations where actors are theoretically rational, but practical irrationality occurs.

Speaking of rationality, Weber presents various different action types: 1) affective, 2) traditional, 3) value-rational, and 4) goal-rational. The first, affective, is driven by emotions such as hatred or love. Affective rationality can also be described as contrasting with instrumental rationality. Traditional actions refer to something that “has always been done,” where actors act with reference to traditional reasoning based on what was done previously. Value-rational actions point towards an action taken because it is in line with a given set of values. Here, it can be argued that the process is valued more than the consequence itself. In contrast to goal-rationality’s means-ends thinking, value-rationality is concerned with acting rightly because it is based on a given value system. Goal-rationality aims at means-end thinking, using the best means to achieve its goal. According to Weber, modernity points towards the first two action types (affective and traditional) becoming less and less prevalent. Value-rational actions are another type of actions. For example, when talking about “rational choice,” it is with goal-rationality as a focal point. Elster argues, on the other hand, that an action is rational based on three optimal decision bases. Firstly, a person must realise their desires, and secondly, their beliefs themselves must be optimal based on the evidence available to the actors. Finally, the person must ensure an optimal amount of evidence or information before the decision is made. According to Elster’s definition of rationality, there is a connection between practical and theoretical rationality. When Elster argues, a person must gather optimal evidence to act rationally. This requires theorising to examine evidence and whether it is relevant to one’s rationality-driven actions.

One popular theory of rationality is that of rational choice theory, but it can be criticised for being a normative theory, formulating what actors should do, rather than what they actually do. The question is whether rational choice theory is more normative than descriptive. When speaking of science and the study of the world, it is vital to make a distinction between descriptive statements that describe what is the case, for example, “I am writing in English,” and normative statements that relate to what should or ought to be the case, such as “I should write in English.” Regarding the distinction between normative and descriptive statements, reference can be made to David Hume’s guillotine, wherein the thesis is presented that an ethical or judgmental conclusion cannot be inferred from purely descriptive factual statements. This differentiates between “is” and “ought.” According to Hume’s guillotine, it is argued that a normative conclusion cannot be derived from descriptive statements. The reason for this is that Hume claimed that descriptive statements describe what is the case in the world but cannot inform about what should be the case. According to Hume, the normative does not follow from the descriptive, and vice versa. Therefore, there must be a necessary logical distinction between “is” and “ought” statements. However, some would defend rational choice theory by arguing that although people do not always act rationally, the models can still have some predictive value in making broad generalisations about society or the economy — however, rational choice’s assumptions can also lead to misguided assumptions and understandings of the research field. Rational choice theory is also quite individualistic. Therefore, according to rational choice, it can be difficult to achieve a complete understanding of actions that do not necessarily appear as self-maximising. Rational choice theory can also be criticised for not incorporating various ethical considerations into actors’ actions. In principle, rational choice theory can be used to rationalise the economic gain from a ruling class suppressing another given minority group.

Rational choice theory is not a theory of science, as there is no clear epistemology and ontology. This theory fundamentally believes that when actors are exposed to various opportunities, they will act rationally. Therefore, the actor will choose the option that maximises their own utility. Rational choice theory can be described as a theory that attempts to describe actors’ actions in social science. However, there are also many criticisms of rational choice theory, as behavioural sciences indicate that people can act irrationally and emotionally, considering the framing effect, for example. In rational choice theory, self-maximisation is a significant concept. Some argue that people are self-maximising, but other theorists also recognize that people can act altruistically. In rational choice theory, it may be relevant to talk about 1) outcomes, 2) different outcome values, and 3) different outcome probabilities. These considerations about outcomes and their value in connection with outcome probabilities can guide making economic decisions. For example, as a farmer, it may be relevant to consider what the outcome, value, and probabilities are in connection with growing wheat versus rye on one’s land area. Rational choice theory typically operates with some idealisations, like there is a complete preference ordering.

Thomas Hobbes thought of rationality in the light of egoistic and calculating individuals who choose to accept a superior power because it offers them a certain degree of security and prosperity. This thinking about rationality is greatly influenced by Hobbes’ conception of humans in the state of nature, which is described as brutal and short-lived. The Enlightenment was based on both the scientific and what can be described as a rational and mechanical view of the world, heavily influenced by scientific positions such as naturalism. Romanticism, which celebrated the emotional bond with the homeland and nature, gradually gave way to the rationalisation of the world brought about by industrialisation. Industrialisation can be described as an entity that largely adopted the Enlightenment project, including its mechanical worldview. In 1776, the British philosopher and economist Adam Smith published “An Inquiry into the Nature and Causes of the Wealth of Nations.” Here, Smith described how the starting point for the economy was actors’ rational self-interest. According to Smith, the division of labour, free trade, and individuals’ self-interest were the basis for creating increasing prosperity. Here, the term “the invisible hand” can be highlighted, which is a metaphor for the unseen forces that drive this market.

Framing suggests that people may not necessarily act in accordance with the “economics 101 rationality”, as the framing of a given issue can influence agents’ choices. Framing effects are empirically well-documented and do not align with rational choice theory. Therefore, it could be argued that the perception of individuals as rational actors may need to be revised. Framing describes how decisions are influenced by the angle from which they are presented. Another case against “textbook rationality”, that is prospect theory, further describes how it hurts more to lose $100 than to gain $100, which can be described as loss aversion. Prospect theory thus challenges the understanding of rationality found in rational choice theory and classical economics. Behavioural economics’ paradigm shift, including prospect theory developed by Daniel Kahneman and Amos Tversky, shows that gains and losses are not equivalent, introducing the concept of “loss aversion,” where the perceived value of loss is higher than the utility gained from an equivalent gain, even though it’s the same amount. What follows from this insight is the complexity of human decision-making and scientific paradigms, according to different epistemologies and worlds, highlighting the need for continuous reevaluation of our understanding of models and theories within social science.

Bibliography

Hume, David (1999). An Enquiry Concerning Human Understanding. Oxford: Oxford University Press.

Kuhn, Thomas (1970). The Structure of Scientific Revolutions (2. udg.). Chicago: The University of Chicago.

Midtgaard, Søren Flinch (red.) og Nedergaard, Peter (red.) (2022). Videnskabsteorien i grundrids. København: DJØF Forlag.

Written by Scott Brodie Forsyth

Danish, English, and History Teacher | Student of Political Science | Blog on Politics, Sociology, and Philosophy