Every methodological choice carries hidden philosophical commitments. The researcher who ignores this doesn’t escape philosophy — they just inherit it unconsciously.
Introduction: Philosophy as Mandatory Equipment
There is a common attitude among doctoral students encountering philosophy of science for the first time: mild impatience, occasionally dressed up as pragmatism. I came here to do research. I want to work with data, build theories, publish papers. Why am I spending time on questions about the nature of reality that were debated by people who died two thousand years ago?
The answer is uncomfortable but inescapable: because you are already doing philosophy, whether you know it or not. Every research question you ask, every construct you define, every method you choose, every causal claim you make — each of these rests on a set of philosophical assumptions about what reality is like and how it can be known. The only choice available to the researcher is whether to make those assumptions consciously and critically, or to inherit them unreflectively from the conventions of the field and the habits of the discipline.
The researcher who never engages with philosophy does not achieve a philosophy-free research practice. They achieve a philosophy-invisible one. And invisible assumptions, precisely because they are never examined, are the most powerful and most limiting constraints on what a researcher can see, ask, and conclude.
This article makes the case for philosophical awareness as a core intellectual competency for researchers in the social sciences and business disciplines — not as an abstract exercise in academic virtue, but as a practical necessity for doing better, more honest, and more genuinely insightful work.
Part I: What Philosophy Actually Is — Ontology and Epistemology
Before addressing why philosophy matters for research, it is worth being precise about what philosophy, in the context of academic inquiry, actually involves. The term is often used loosely, and its looseness breeds the kind of dismissiveness that keeps researchers from taking it seriously.
At its core, philosophy of research is concerned with two fundamental questions.
Ontology: What Is the World Like?
Ontology is the branch of philosophy concerned with the nature of reality. The ontological question is: what exists, and what is the character of its existence? Is reality objective — does it exist independently of any observer, fixed and waiting to be discovered? Or is reality socially constructed — does it come into being through the interpretations, agreements, and practices of the people who inhabit it? Or is the answer somewhere more nuanced, with different kinds of reality requiring different kinds of ontological commitment?
These are not idle questions. Consider something as mundane as a bottle of water sitting on a table. Does that object exist as an independent physical fact about the world? Yes, in one obvious sense — the molecules are there, the physical properties are measurable. But does it exist as a bottle of water, specifically? That designation depends on a set of socially shared concepts and categories — the concept of a bottle, the concept of water as a drinkable substance, the social practice of packaging liquids for consumption. Strip away the socialized consensus and the physical object remains, but the bottle of water disappears. Its existence as a bottle of water is partly a function of the world and partly a function of the conceptual framework that human beings have developed to organize their experience of the world.
This seemingly abstract point has immediate, concrete consequences for research. When researchers decide what to study — which entities to treat as real, stable objects with measurable properties — they are making ontological commitments. Those commitments shape everything that follows.
Epistemology: How Do We Know?
Epistemology is the branch of philosophy concerned with knowledge. The epistemological question is: what counts as knowledge? How do we evaluate whether a claim about the world is true, justified, or reliable? What methods are capable of generating knowledge, and under what conditions?
In research practice, epistemology is made concrete through methodology. When a researcher chooses a particular method — a controlled experiment, a survey instrument, an ethnographic interview, an archival dataset — they are making an epistemological bet. They are asserting, implicitly or explicitly, that this method is capable of generating reliable knowledge about the phenomenon they are studying. The method embodies a theory of how knowledge is produced.
But methodology does not float free of deeper epistemological assumptions. The choice to use a regression model to test a causal hypothesis assumes that causal relationships between variables can be reliably inferred from patterns in quantitative data. The choice to conduct in-depth interviews assumes that human meaning and experience — which are not directly observable — are nonetheless accessible through language and reflection, and that this accessibility can yield genuine insight into social phenomena. Neither assumption is obviously correct or incorrect. Both are philosophical commitments that deserve examination.
The Relationship Between the Two
Ontology and epistemology are deeply intertwined. What you think the world is like shapes what you think can be known about it and how. A researcher who believes social reality is objective and measurable will be drawn toward methods that produce quantifiable evidence — because on that ontological view, quantification is the appropriate epistemological tool. A researcher who believes social reality is constructed through meaning and interpretation will be drawn toward methods that illuminate those meanings — because on that ontological view, statistical patterns in large datasets may miss the most important things.
The point is not that one ontology is right and the other wrong. The point is that the choice has consequences, and that those consequences need to be understood and owned by the researcher who makes them.
Part II: The Self-Reinforcing Cage — How Concepts Trap Thinking
The Lock-In Problem
One of the most useful concepts for understanding why philosophical awareness matters in research is what might be called the self-reinforcing cage. The image is apt: research traditions develop concepts, and those concepts become the lens through which subsequent researchers see the world. New work builds on existing concepts. Citations accumulate. The concepts acquire the weight of consensus. And gradually, without any single decision being made, the concepts stop being questioned. They become background assumptions — part of the furniture of the intellectual world, taken for granted rather than examined.
This is not unique to any particular field or tradition. It is a structural feature of how academic knowledge production works. Every discipline is, to some degree, a community of scholars who share a set of conceptual frameworks. That sharing is what makes communication and cumulative progress possible — you cannot build on prior work if you cannot assume that your audience shares your basic vocabulary. But the same feature that makes productive conversation possible also creates intellectual lock-in. The more a concept is used, cited, and built upon, the higher the cost of questioning it. Doing so means stepping outside the shared vocabulary, challenging assumptions that the entire conversation has been predicated on, and risking the charge of not understanding the field.
The result is that many concepts in academic disciplines are deployed for years or decades without being subjected to the philosophical scrutiny that would either confirm their validity or reveal their limitations. They become what might be called closed concepts — constructs that circulate freely in the literature without anyone asking whether they are actually the right concepts for the phenomena being studied.
The Cost of Closed Concepts
The cost of closed concepts is not merely philosophical tidiness. It is a practical limitation on the quality and reach of research. When researchers adopt concepts unreflectively, they foreclose entire lines of inquiry. The questions that closed concepts make unaskable may be precisely the questions that matter most.
Consider how concepts in Information Systems research have at times treated technology as a stable, well-defined object with fixed properties that cause predictable outcomes in organizations. This framing has generated a substantial body of useful research. But it has also — precisely because it has been so widely adopted — made certain questions difficult to ask. If technology is a stable object that causes outcomes, then the question of how technology comes to be what it is, how it is shaped by the organizational, social, and political processes that produce and implement it, how it changes in response to use — all of these questions are systematically backgrounded. The framing has built-in boundaries, and researchers working within it often don’t notice those boundaries because the framing itself makes them invisible.
Philosophical thinking is the practice of making those boundaries visible. It asks: what are the concepts I am using? Where did they come from? What do they assume? What do they preclude? Are these the right concepts for the phenomenon I am trying to understand, or have I inherited them from convention without examining whether they fit?
Part III: The Positivist Cage — “Technology Causes X”
Unpacking a Common Research Statement
One of the most instructive ways to see the philosophical assumptions embedded in ordinary research practice is to examine a statement that appears constantly in the business and information systems literature: technology causes X. This statement — or more specific versions of it, like “AI adoption increases productivity” or “social media use affects well-being” — is the foundation of a huge proportion of empirical research. It looks like a simple empirical claim. It is actually a dense package of philosophical commitments.
Assumption One: Subject-Object Separation
When you say “technology causes X,” you are assuming that technology and the human beings who use it are distinct, separable entities. Technology is the cause; human behavior or organizational outcomes are the effect. The causal arrow runs in one direction, from an independent entity to a dependent one.
This is a positivist ontological assumption. And for many research purposes, it is a reasonable and productive one. But it rules out, from the outset, the possibility that the relationship between technology and humans is constitutive rather than causal — that technology and human practice are mutually shaped, that the technology becomes what it is through the practices that produce and use it, and that human practice is shaped by the material properties of the technology in ways that cannot be cleanly decomposed into causes and effects. The sociotechnical research tradition, by contrast, starts from precisely this assumption of mutual constitution. It asks not what technology does to humans but how technology and human practice co-evolve and co-constitute each other.
Neither approach is simply right or wrong. But they generate fundamentally different research questions, and a researcher who has never examined the philosophical assumptions of the first approach does not know that the second exists as an option.
Assumption Two: Technology as Stable Entity
Saying “technology causes X” also assumes that technology is a sufficiently stable, well-defined entity to function as a meaningful independent variable. You can measure the presence or absence of the technology, or its characteristics, and treat that measurement as capturing something real and consistent across contexts.
But is technology stable in this way? In many contexts, what a technology is — what it does, what it means, how it functions — varies substantially across the organizations, teams, and individuals who use it. The same enterprise software system becomes, in practice, a very different technology in a firm that uses it extensively and creatively versus a firm that uses it minimally and with resistance. The technology does not have a fixed set of properties that cause fixed outcomes. It has properties that interact with organizational culture, management practice, employee skill, and institutional context to produce outcomes that are highly contingent.
This is not a minor technical complication. It is a philosophical issue about the nature of the entity being studied. And a researcher who asks “does technology X cause outcome Y?” without asking “what is technology X, actually, in these different contexts?” may be generating results that are technically valid within a narrow set of assumptions and practically misleading about the phenomenon they purport to explain.
Assumption Three: The History of Technology Is Irrelevant
A third assumption embedded in “technology causes X” is that what matters about the technology is its current state. The causal claim is synchronic — it treats the technology as it exists at the moment of observation, and asks what it does. What is not asked is how the technology came to be what it is: the design choices that were made, the interests they reflected, the alternatives that were foreclosed, the organizational processes that shaped the implementation.
In many research contexts, the history of a technology is not just irrelevant background — it is the key to understanding why the technology has the effects it does. A system that was designed with efficiency as the primary value will embed different assumptions about work and workers than a system designed with user experience as the primary value, even if both are nominally “the same type of technology.” The philosophical choice to treat technology as a stable causal agent, rather than as the product of a history worth investigating, eliminates this entire dimension of inquiry.
Part IV: The Typing Example — One Phenomenon, Two Philosophies, Two Realities
Perhaps the most vivid illustration of how philosophical assumptions shape research conclusions is a comparison drawn from the philosophy of technology: the act of typing, viewed through two radically different philosophical frameworks.
The Simon View: Mechanical Efficiency
From the perspective of scientific rationality associated with Herbert Simon and the tradition of classical AI and cognitive science, typing is a mechanical process. The typist is an information-processing system. Typing skill is the internalization of rules — the efficient mapping of intended outputs to motor commands. The purpose of typing is instrumental: to produce text as efficiently and accurately as possible. Agency in this framework is the capacity to execute rule-governed processes reliably.
On this view, the ideal typing assistant — or, in the contemporary context, the ideal AI support for typing-related tasks — is one that maximizes speed, minimizes errors, and reduces the cognitive load associated with the mechanical aspects of text production. The design criteria are efficiency, accuracy, and compliance. Typing is a means to an end, and the technology that supports it should optimize that means.
The Heidegger View: Embodied Practice
From the perspective of Heideggerian phenomenology and the tradition of interpretive philosophy of technology, typing is not a mechanical process but an embodied practice embedded in a web of meanings. The typist is not an information processor but an actor whose activity is constitutive of identity. Typing skill is not rule-following but a form of embodied, situated competence — the kind of knowing that lives in the hands and cannot be fully articulated in propositional terms. The context of typing — what is being written, for whom, to what end, within what relationship — is not incidental background to the mechanical act but the substance of what the act means.
On this view, identity and engagement are inseparable from typing practice. The writer who is deeply absorbed in composing a letter is not just efficiently producing keystrokes; they are enacting who they are in relation to another person, to a problem, to a body of ideas. The act shapes the actor.
Two Designs, Two Worlds
These two philosophical frameworks lead to entirely different design conclusions for a typing assistant. The Simon-inspired design focuses on efficiency: autocomplete, error correction, speed enhancement, reduction of cognitive friction. The Heideggerian-inspired design asks a different set of questions: does this tool support or undermine the writer’s sense of engagement and identity? Does it replace the writer’s voice with something generic? Does it create disengagement by automating the parts of writing that are most meaningful? Does efficiency optimization come at the cost of the depth and authenticity that make writing worthwhile?
Both sets of questions are legitimate. But a designer who has never examined their philosophical assumptions will default to the Simon framework — not because it is better, but because it is the framework that the dominant traditions of technology design have inherited and normalized. The Heideggerian questions will simply not occur to them, because they are not visible from inside the cage.
This example makes concrete what is otherwise easy to treat as abstract: philosophical assumptions have direct, practical consequences for what gets built, who benefits, and what gets lost.
Part V: Deliberate Philosophy Versus Inherited Philosophy
You Always Have a Philosophy
The central practical takeaway from all of this is simple to state and difficult to internalize: you always have a philosophy. Every research decision embeds philosophical commitments. The question is not whether to have a philosophy but whether you have chosen it deliberately or received it by default.
Inherited philosophy is the philosophy of the unreflective researcher — the one who uses the concepts the field provides without asking where they came from, who adopts the methods that are standard without asking what they assume, who makes causal claims without asking what model of causation those claims depend on. This researcher is not philosophically neutral. They are philosophically dependent — their intellectual framework is not their own, in the sense that it has never been examined and owned. They are, in the image introduced earlier, inside a cage they cannot see.
Deliberate philosophy is not the same as philosophical relativism. It is not the claim that all frameworks are equally valid and the choice between them is arbitrary. It is the practice of asking, for each significant choice in a research project: what are the philosophical assumptions behind this? Are those assumptions appropriate for the phenomenon I am studying? What do they enable and what do they preclude? Would a different set of assumptions lead me to ask more important questions or generate more useful knowledge?
This is not a once-and-done exercise. It is an ongoing practice of intellectual self-examination — maintaining what might be called a philosophical antenna, a sensitivity to the assumptions that are always present in research work, and the humility to ask whether those assumptions deserve the confidence they are usually accorded.
Philosophical Humility as a Research Virtue
There is a kind of researcher confidence that is actually a form of philosophical incuriosity — the confidence that comes from not having asked the hard questions. The researcher who has examined their assumptions may feel less certain in some respects: they know that the concepts they are using are human constructions, not natural kinds; they know that their methodology is one approach among several defensible alternatives; they know that their findings are conditioned by choices they made that could have been made differently.
But this epistemic humility is not weakness. It is intellectual honesty. It produces research that is more carefully qualified, more alert to alternative interpretations, and more genuinely open to the possibility that the most interesting questions are not the ones the existing framework makes easy to ask.
The researcher who has engaged seriously with philosophy does not abandon rigor. They bring rigorous thinking to bear on the assumptions that less reflective researchers treat as settled. This is harder work, but it is also more honest work — and, in the long run, more productive work, because it keeps open the possibility of genuine intellectual progress rather than the mere accumulation of results within a fixed and unexamined framework.
Part VI: Data Does Not Speak for Itself
One particularly important instance of inherited philosophical assumption in contemporary research deserves separate attention: the widespread belief that patterns in data are self-evidently meaningful — that sufficiently large datasets, analyzed with sufficiently sophisticated methods, will reveal truths about the world without the need for theoretical interpretation.
This belief — sometimes explicit, more often implicit — rests on a set of philosophical assumptions that are rarely stated and almost never examined. It assumes that data is a direct representation of reality rather than a constructed artifact shaped by the choices of whoever collected it and defined its categories. It assumes that correlation is a reliable guide to understanding, and that understanding phenomena can be equated with predicting them. It assumes that the patterns found in data have significance that is independent of the theoretical framework used to interpret them.
None of these assumptions is obviously true. Data does not speak for itself. It speaks through the concepts and frameworks that researchers use to collect, organize, and interpret it. The patterns that emerge from a dataset are always already shaped by the decisions about what to measure, how to measure it, what to include and exclude, and what questions to ask of it. Two researchers with different philosophical commitments, analyzing the same dataset, may reach entirely different conclusions — not because one is wrong and the other right, but because their interpretive frameworks lead them to see different things in the same data.
This is not an argument against quantitative research or large-scale data analysis. It is an argument for bringing philosophical awareness to that work — for recognizing that data analysis is always theory-laden, that the patterns that surface in data only become findings when interpreted through a conceptual framework, and that the choice of framework is a philosophical act that deserves conscious attention.
The researcher who believes data speaks for itself is not being rigorous. They are being philosophically naive in a way that may, ironically, undermine the rigor they believe they are achieving.
Conclusion: Philosophy as the Foundation, Not the Decoration
The argument of this article is not that every researcher needs to become a philosopher. It is that every researcher is already doing philosophy, and that the quality of their research depends, in part, on how well they do it.
Philosophy is not the decoration that gets added to research to satisfy doctoral program requirements. It is the foundation on which every research decision rests. The concepts researchers use, the methods they choose, the causal claims they make, the findings they report — all of these are shaped by philosophical commitments that can be either examined or ignored, but cannot be avoided.
The researcher who examines those commitments gains something concrete: the ability to ask whether their framework is appropriate for their phenomenon, to recognize when inherited concepts are limiting their vision, to see the questions that their methodology makes invisible, and to understand why different researchers looking at the same phenomenon reach different conclusions. They also gain something less tangible but equally important: intellectual humility — the recognition that knowledge is hard-won, always conditioned, and never final.
The cage is always there. Philosophy is the practice of knowing you are in it — and occasionally, finding the door.
This article is based on a doctoral seminar discussion on the philosophy of research, ontology, epistemology, and the role of philosophical awareness in rigorous academic inquiry.
Leave a Reply