Introduction
This book project started as a frustration with the modern state of social science research. Researchers tend to associate themselves with one of two camps, usually arguing strongly in favour of their own approach. One side considers themselves to be qualitative researchers interested in the context, stories and experiences of research subjects, whereas the other side, the quantitative researchers, are interested in interrogating objective facts and finding statistical patterns with graspable numbers. Motivated by student’s wishes to be well prepared for the job market, many department programs maintain this division in an attempt to provide enough training for both sides to appeal to potential employers with an array of concrete skills and practical tools at hand. And who could blame them? The university is under constant attack by budget-constrained (and constraining) politicians, the cost of living grows ever more, and the concomitant pressure on household budgets makes the question of a return on higher education a prescient one. To complicate matters more, these pressure factors intersect with the democratization of knowledge access through the internet and the yet to be understood transformation that AI will bring to how knowledge, teaching and work will happen in the coming years. All these uncertainties make falling back on tried and tested skills an enticing and arguably reasonable proposition. And certainly, contrasted with the abstractness of Critical Thinking and reflexivity, the idea of developing hard labour market skills is a comforting notion.
This is where we run into an obstacle - after all, how do we assess which skills are right for the labour market? In the past, a newspaper typesetter and a radio repair person had seemingly comfortable and relevant sets of skills until those skills became superfluous to the labour market. Yet, unlike the newspaper typesetter, whose system of knowledge became unnecessary, the radio repair person’s system of circuits and electronics did not change, even if the specific types of things that needed repair did. In other words, the radio repair person was trained in a system that persisted despite their specific training. The question in the social sciences today, therefore, is whether the skills we are teaching are those that are persistent because they are systemic, or those that will become superfluous?
Ideally, as social scientists, we gain a systemic approach to understanding the complex world we live in, which helps us derive the necessary methods to understand that world properly. However, in our current higher education syllabi, we often still focus on a methods-first approach that maintains a divided methods paradigm, despite a significant literature pointing out the limitations of this approach. This book offers to reframe this problem not in terms of debates about techniques and tools of data analysis. In fact, this book argues that the disagreements that run deepest in social research are not disagreements about evidence or about technique, but about the discipline’s assumption about what social reality is. Every discipline, and its associated methods, disagrees with the others about what they society to be, and these assumptions are almost never explicitly stated. Such a clarification would not only foster interdisciplinarity as a real, viable option, but it also affects the quality of work of each individual researcher. After all, scientists who are aware and can state their own assumptions do better work than researchers who cannot: they design more coherent projects, read other people’s work more accurately, and know what their findings are incapable of showing. Everything that follows is an attempt to make good on the claim that our assumptions about social reality can be known and made explicit and to hand you the tools to act on it, for better, genuinely interdisciplinary science.
I. Four Rigorous Reports, Four Different Cities
To understand what we mean when we say different ideas of social reality, let us entertain an example for the sake of illustration.
Imagine a committee convened to advise a mid-sized European city on why poverty in three of its districts has stayed at the same levels for twenty years. The city is very thorough and, commendably so, commissions economists, sociologists, political scientists and urban geographers. Every team provides a reports by the deadline, which is already impressive, and every report is of the highest quality.
However, as you might imagine, as the outcomes are presented, wildly different analyses of the problem become apparent. First, the classically trained economists explain that poverty persists because of skill deficits and a labour market that rewards neither training nor low-wage work; they recommend vocational programmes and specific activation policies. The next ones to present are the sociologists, who explain that poverty persists due to residential segregation, network disadvantage and the slow accumulation of institutional exclusion; thus, as a response, they recommend neighbourhood investment and the structural reform of welfare institutions. Then the political scientists explain that redistributive organisations have been dismantled across three decades of fiscal retrenchment; so their recommendation points towards political organisation and electoral strategy. Finally, the urban geographers present, and by this point you realise that they too will provide an entirely different explanation. Indeed, they present their findings that poverty in the three districts is reproduced through housing market systems and historic area-specific policies have over time led to socio-economic segregation; they recommend specific area-based interventions and social housing reforms.
Not one of these reports is wrong. Each is based on decades of careful empirical work, each is internally rigorous, and each was written by people who know their field. The committee, which began the process short of ideas, now has four of them and no way to choose. It cannot act on all four at once, and it cannot rank them, because the reports are not comparable. This is because, unlike the first impression on might gain, these are not four views of one object. Instead, they are four claims about what kind of object poverty is.
Now we see that the incompatibility runs deeper than disciplinary habit, and it becomes particularly visible the moment one asks what each report takes to be doing the causing. What are the objects, agents or entities that are actually doing something and affecting things? The answer to such a vague and yet uncomfortably to the point question, varies of course: for the economists, causation runs through individual decisions taken under certain constraints; the sociologists would argue that causes are passed through relational position and structural locations; for the political scientists, causation becomes clear through organised interests and institutional rules; whereas the urban geographers, find causation running through spatial form and material infrastructure. Causation is an excellent example to inflame the dissenting social scientist, because a causal claim is the point at which research commits itself, even if this often goes unnoticed. To say what produces an outcome, is to say what kinds of things exist and are capable of producing anything at all. So, these four accounts are not four levels of a single process, waiting to be stacked into a comprehensive model. Indeed, even if one were to step away from this toy example and consider for example other, non-mainstream, heterodox approaches to economics, many persistent disciplinary ontological assumptions persist, as almost all economic approaches still focus on an aggregative understanding of individuals rather than systematically integrating groups. Thus, these disciplines, and to a lesser extent also within the disciplines, there are competing claims about what is causally primary, about what populates the social world, and about the scale at which explanation should operate. This means the different social scientists of our example did not pick different units of analysis (despite studying the same thing) because they wish to vex a city council, but because their conception of what the nature of the thing that they are studying, in this case urban poverty, is fundamentally, philosophically different. Hence, assembling these disciplines without confronting those fundamental differences produces not a synthesis but merely a painfully long document.
Assumptions of this kind, about what exists, what it is made of and what it is able to do, are ontological assumptions, and the study of them is called ontology. Every research project has them. Every researcher has them. Even you! However, very few researchers or projects state them. They are absorbed during training, carried in the standard moves of a discipline, and reproduced in the choice of what to measure and what to leave as background. Because they hide covertly behind the argument rather than being explicitly in it, they cannot be defended, contested or improved. This is also what we see in our example: what our committee lacks is neither data, nor method. Instead, it lacks a vocabulary that can cater to the differences that actually divide it.
The most visible expression of that currently still predominant lack is the divide between quantitative and qualitative research, usually described as a disagreement about technique: counting against interpreting, measuring against understanding. This description is inadequate, and we will problematise it properly later in the book; for now we can take the divide as researchers themselves experience it and look at what sits underneath. One side proceeds as though social phenomena are regular, and similar enough to natural objects, that statistical analysis can disclose their structure. The other side proceeds as though social phenomena are constituted through meaning and interpretation in ways that make aggregation inadequate to what they are (Bryman, 1988; Goertz & Mahoney, 2012).1 Both positions are defensible, yet these fundamental critiques are so rarely explicitly discussed, that instead, the central argument is conducted in the currency of method, where it cannot be settled, because it is not really an argument about method! Yet, we would like to emphasise that this is not a failure of intelligence or good faith on anybody’s part. Most researchers were never given the concepts and the vocabulary that stating such a position requires nor systematically required to do so. Supplying this vocabulary is what this book attempts to do.
The way disciplines are organised only compounds the difficulty, since training in economics and training in sociology introduce not only different questions but set different standards of evidence, different criteria of adequate explanation and different instincts about what lies outside the field’s proper scope. But these conventions are not arbitrary, in fact, they encode decades of accumulated judgement about what matters. But they carry ontological commitments that silently (but with great force!) limit what any individual researcher is able to see, and the interdisciplinary work that might expand those limits, runs into the issue of the same missing vocabulary. This is why the best interdisciplinary research that exists in the various fields, usually handle this rather well. Such work often shows an intuitive grasp of where its constituent frameworks stop agreeing, and of which one to follow when they do not. However, it rarely makes that understanding explicit, which means the achievement remains personal to its authors: admirable, instructive and almost impossible to teach or to reproduce systematically. A formidable example is to do so, is presented by Barry and Born (2013), who come closest to naming what is at stake, which is identifying a logic of interdisciplinarity that works by transforming the objects of knowledge rather than by combining existing disciplines or subordinating one to another. Turning that tacit skill into something learnable is a large part of what we are attempting here.
However, the importance of this knowledge is not merely to beef up the methods section of future papers, but to make possible a social science that is equipped for this world, and the stakes are high. Entrenched inequality, democratic instability, climate mitigation and adaptation, forced migration and the governance of emerging technologies are problems whose treatment depends on understanding social mechanisms well enough to intervene in them. That understanding requires frameworks that can tackle the complexity of social life rather than the conventions of disciplines organised a century ago for other purposes. If the social sciences cannot build cumulative, integrative knowledge across their internal divisions, the cost is not only academic untidiness but practical inadequacy at precisely the moment when adequate social science matters most.
Finally, this book began in a classroom. It grew out of courses taught at Erasmus University College in Rotterdam, where students receive disciplinary training in economics, sociology, political science, psychology and the humanities. In an attempt to provide methods training that enables effective interdisciplinarity, this book emerged, almost by necessity, and improved over time. Teaching this course over several iterations taught us what is genuinely hard about this material, which concepts have to be established before others make sense, and what turns the framework from something students can describe into something they can use. The sequence of what follows is a product of that teaching, and a great deal of correction and clarification was possible due the students themselves. Thus, what we offer is not a set of conclusions to be accepted, but a set of tools, as presented in the next section.
II. The Approach: From Implicit Assumptions to a Shared Vocabulary
Having now been faced with a considerable problem, let us consider how a comparable issue is handled elsewhere. Twentieth-century physics produced two spectacularly successful theories. General relativity describes gravitation as the curvature of spacetime and performs superbly at large scales. Quantum field theory describes matter and the remaining forces through discrete, probabilistic interactions and performs superbly at small ones. Both are confirmed to extraordinary precision. They are also mutually inconsistent: their accounts of space, time and causal determination cannot both be literally true, and where their domains overlap, in the interior of a black hole or the first instants of the universe, combining them yields nonsense. Physicists do not respond by declaring one field unscientific or by agreeing to a tactful silence. They treat the inconsistency as a central research problem of the discipline, they can say precisely where the two frameworks contradict one another, and much of the past half century of theoretical physics consists of attempts to build something that neither theory can supply alone (Rovelli, 2004).
That is what a field can do once it possesses a shared apparatus for locating its own foundational disagreements. The social sciences are not in that position, and it would be naive to propose that they simply copy the arrangement, because our (social) objects are of a different kind. Physical processes hold no theories about themselves, do not read what we write about them, and do not behave differently because a description has become fashionable. Social reality includes people who interpret their own situation, disagree with one another about what it is, and remake it in the light of those interpretations (Winch, 1958; Giddens, 1984), a circumstance that significantly complicates the matter. Decisive experiments are rare for us, and the closure a laboratory could provide is mostly out of reach, because social systems remain open. That openness is not a defect to be engineered away but a feature of what social reality is (Bhaskar, 1979; Lawson, 1997). In fact, attempts to arrange social science as though closure were available are instructive in their own right, since they show how much needs to be assumed away to produce it. This is exactly why the work of specifying where frameworks diverge has to be done conceptually rather than left for evidence to settle. The comparison with physics is helpful, although not because it hands us instructions to copy, but because it shows what becomes possible once a field can name its own foundational disputes.
The response this book proposes follows from that. If fragmentation between disciplines continues to exist because ontological commitments stay implicit, then the work that must be done is to bring them into the open, examine them, and build a vocabulary adequate to the differences between them. This is not an attempt to install a single ontology across the social sciences or to talk their differences away. Some of those differences are real disagreements about the nature of social reality and will not dissolve under better terminology. But researchers who know what they disagree about can learn from one another, borrow carefully, and understand where integration is available and where it is not. Researchers who do not know what they disagree about talk past each other, and as you may have experienced, frequently at length and rarely pleasantly.
The framework organises these commitments into four debates. They are not an arbitrary selection, but mark the points at which background assumptions about social reality diverge most consequentially, and they recur, under various names across history, wherever social scientists argue about foundations. Each is best understood as a question about where a position sits between two poles and how it considers the interaction of the two sides, rather than as a choice between them.
The individuals and groups debate asks where causal power is located: how much explanatory work is done by individuals and their interactions, and how much by collectivities, institutions and structures whose emergent properties and causal powers are not reducible to their members. The structure and agency debate asks how the patterns that precede any particular action relate to the actions through which people reproduce, transform or resist the social conditions around them, and therefore what is treated as durable, what as open, and by what mechanism change occurs. The material and ideational debate asks how material interests, economic relations and physical resources stand in relation to meanings, norms, beliefs and discursive structures. The transcendence and immanence debate asks a twofold question of specificity and overarchingness; once in regards to if and how there are universal features of society, or if each society is entirely specific to its space-time constellation; the second asks about conceptions of causality, in terms of conceiving the causes for a specific social phenomenon to exist within it, or beyond it.
Two things must be noted as we briefly introduce the debates this way. The first is that very few researchers occupy a pole; most hold some middle position, and there are plentiful middle positions available. What distinguishes them is not where they sit on a line but how they theorise the interaction between the two sides, which is a far more interesting question. The second is that every research project takes positions on all four, whether or not its author has ever considered them. Those positions necessarily dictate what the researcher is able to see, which explanations are even within the realm of the thinkable, and which methods can reach the phenomenon as conceived. Making these ontological assumptions explicit is the first step toward a design that is coherent and defensible rather than arbitrary and conventional (however enticing that might seem).
A researcher equipped with this framework can do several things that specialist training alone does not supply. We hope that equipped with these tools and the ability to situate any discipline, she can read a study and place it, not only by topic and technique but by the ontological position it occupies and the explanatory consequences that follow from it. She can tell the difference between two accounts that examine different aspects of the same thing and two accounts that disagree about what that thing is. Consider for example, three studies that explain the long decline in electoral turnout. The first attributes it to a weakening sense among voters that voting makes any difference; the second to the fact that parties now contact far fewer people directly. These differ in what they chose to look at, not in what they take turnout to be: both treat it as the sum of individual decisions taken under particular conditions, and their findings can be added together. The third, however, explains the decline through the erosion of the associational fabric, the unions, churches and local party branches through which people were once drawn into political life, treating these as structures whose organising capacity is not reducible to the dispositions of their members. That account in contrast to the first two, is not addressing the same object, and no quantity of additional survey data will reconcile them until the prior question of what turnout is, or what it is a property of, has been answered. The first difference is thematic. The second is ontological, and only this second difference needs to be clarified before the evidence can be understood on equal grounds. Furthermore, with the approach proposed in this book, a social scientist can design her own research with the coherence that comes from knowing what she is assuming, rather than importing incompatible assumptions from three literatures at once and, in the best case, discovering the problem somewhere in the discussion section. She can argue across disciplinary and paradigmatic lines with the precision that real intellectual exchange requires. And she can hold her own framework at arm’s length, to be critically considered, which is the difference between using a theoretical lens and living inside one.
Methods are where all of this becomes concrete, and they matter to this project more than the philosophical vocabulary might suggest. After all, we argue that a method is not a neutral instrument to be picked up by anyone for any purpose. A regression model, a life-history interview, a comparative case design, an ethnography and an agent-based simulation each presuppose something about what social entities are, how they relate, and what we can know about them through what data. Choosing a method is therefore an ontological act, performed in most research without comment, nor awareness. Part III takes the major methods of the social sciences in turn and asks what each assumes, what it can consequently reach, and what it necessarily leaves outside the frame. The purpose is not to rank methods but to let researchers match them to the object under investigation rather than to the habits of a discipline.
The payoff extends well past the individual project. When researchers can state their ontological positions precisely, interdisciplinary work becomes something more stable and reproducible than trust in the collaborator. If they are aware of their positions, the economist and the sociologist working on inequality can identify exactly where they part company: which units are treated as causally primary, which mechanisms are taken to be doing the work, or the temporal and spatial scope the explanation is meant to cover. Naming the divergence guarantees no agreement, but it converts a conversation held at cross purposes into one that can make progress. It also often produces a result that surprises people at first. Two researchers from different disciplines who hold similar ontological positions often collaborate far more easily than two researchers from the same discipline who do not. Shared foundations about what society is, travel across fields much better than shared subject matter travels across ontological assumptions, and much of what is difficult about interdisciplinarity turns out to be ontological rather than discipline. That is good news, because ontological differences can be identified, discussed and sometimes resolved. This book is written in the conviction that there is a real possibility to develop a shared language that can foster interdisciplinarity as a sustainable project.
III. Why Is There So Much Philosophy in This Book?
If you came here for a research methods textbook and find yourself several pages into ontology with no sampling strategy in sight, then a question might be gnawing away at you: why not go straight to research design, methods and substantive social theory, and leave philosophy to the philosophers?
Unfortunately, the answer is that there is no straight route. Every decision in research design rests on philosophical commitments, and the only question is whether they are made knowingly. Once you choose a method, you make a claim about what kind of access to social reality that method provides. When you formulate a research problem, you make a claim about what even counts as a problem worth investigating and what would count as having resolved it. Even the seemingly reflected task of building a theoretical framework is still making a claim about what kinds of entities and mechanisms exist in the domain that often happens unconsciously. These claims are unavoidable; they are made in every project, including the ones that would consider themselves resolutely atheoretical, and making them well requires the vocabulary to express them explicitly. That vocabulary is philosophical, which may indeed be an inconvenience to the unenchanted, but will prove to be rather useful upon the emergence on the other end.
Also, let us specify that the philosophy in this book is not here just for the delight of an interesting argument (although perhaps some may share such a sentiment), and that it is also not an invitation to abandon fieldwork for conceptual analysis. It is important to note that philosophy cannot settle questions that only empirical investigation can clarify, and we will not pretend otherwise. The claim we argue for is more specific and, we think, more useful: research conducted without philosophical self-awareness tends to reproduce the limits of whichever tradition trained the researcher, while also lacking the means to recognise those very limits as limits or to see what a more adequate framework would require. However, when the two are paired, the depth of the research can be vastly improved, which can feed back into theoretical developments that can in turn improve the research and knowledge of certain phenomena! Thus, every philosophical move in this book is made in the service of empirical work, and we have tried to cut whatever does not serve that end.
IV. An Unfinished Book
This textbook is an ongoing project rather than a finished system, and it is deliberately open about the fact that it was developed and made possible due to a collaboration with an LLM. Rather than presenting a theoretical apparatus that claims comprehensive and final knowledge, we position the book as something that will keep developing through dialogue with readers, practitioners and the wider research community. That positioning follows from a commitment about the nature of knowledge (as will be expanded on throughout the book), and from the practical recognition that ontological analysis itself evolves through engagement with research practice.
There is a further reason for making, and keeping, the book open. Knowledge is necessarily partial. Research is always conducted about something that exceeds any particular account of it, an object that is never quite what any of us had in mind and yet is what all of us are talking about. Disciplines approach that object from different positions and report what can be seen from where they stand. What this book offers is not a vantage point from which the object finally appears whole, since humans always have a particular perspective, but a more organised way of speaking about the object: a means of establishing which part of it each account can speak to, and of assembling those accounts without mistaking any one of them for the thing itself (Bhaskar, 1975).
The book was written with substantial assistance from artificial intelligence, and was only with this assistance made possible to come into being within the timeframe we were working with, as having been developed as the complementary text to the course in an earlier iteration within the course of one year. In terms of the collaboration, we ought to clarify the roles taken upon by the collaborators: the philosophical commitments, the analytical architecture and every evaluative judgement in the book are human, and they emerged from a dialogue in which human scholars supplied the foundations, the criteria and the theoretical decisions that guided what the collaboration produced. What the collaboration added was iteration, by making it possible to cycle through far more versions than when handwriting every paragraph. The framework was built from scratch and then rebuilt, reframed, edited and rewritten many times over, with each version tested against the question of whether it actually helped anyone think. A great deal of that testing was done by students, whose confusions proved more informative than their agreement and whose objections are responsible for several of the distinctions that follow. The result is not a system handed down, but an approach worked out, and its adequacy, conceptual as much as didactic, is still improving. We ask that the arguments be judged on their coherence, their analytical sophistication and their practical utility rather than on assumptions about how they were generated. The invitation to disagree with us is sincere: ontological frameworks and theory develop through use and argument, and this one is no exception. Where you find it failing, we would like to hear about it.
This is also why the book is published where it is. Rather than a fixed edition revised every few years, the text is hosted on confused4now.org, an open-access platform for academic literature on which readers can comment on and edit the material directly, with suggestions moderated by the authors and the full version history of every chapter openly accessible. The arrangement follows from the argument. If knowledge is partial and revisable, the form in which it is published should be able to show it: readers can see how and why a chapter changed, students can return years after their course to the most developed version of what they studied while still reading the text they were taught from, and objections that would once have disappeared into a seminar discussion become part of the record. Students who edit the text while taking the course practise scientific debate and peer review with something real at stake, and a wider readership brings the cognitive diversity that produces the objections we would not have arrived at ourselves. So, as you are reading and dislike any point or argument of the book, you can say so on the page itself, or even the section itself.
V. How the Book Is Organised
The book moves through three parts, starting with philosophical foundations, then moving to analytical tools to situate ontologically, and then finally moves on to the practice of research design.
Part I establishes the philosophical and conceptual foundations of ontological analysis. Chapter 1 introduces ontological analysis in social research, tracing its philosophical roots from the empiricist challenge and the Kantian response to the debate between realism and idealism that frames the approach taken here. Chapter 2 develops the conceptual architecture underlying the book as a whole: the relations between ontology, epistemology, methodology and methods; the philosophical treatment of causality from Descartes through Hume and Kant to contemporary debates; and three approaches to emergence and collective causal powers. Chapter 3 addresses stratified reality and the problem of unobservables in social research, developing Bhaskar’s three domains of the empirical, the actual and the real, and introducing the four debates that structure the social sciences. Chapter 4 prepares the analytical ground for Part II by introducing the object of research as the referent that governs the design process, developing the logic of ontological situating and establishing the heuristic character of that practice.
Part II examines the four debates in sustained detail and builds the capability to locate theoretical positions and judge their explanatory consequences. Chapter 5 addresses individuals and groups, Chapter 6 structure and agency, Chapter 7 the material and the ideational, and Chapter 8 transcendence and immanence. Each chapter traces the philosophical depth of the debate, maps the continuum of positions between its poles, and develops diagnostic questions for locating any given theoretical framework along that continuum.
Part III applies the foundations and the tools to the practice of research design. Chapter 9 extends ontological analysis to time, space and reduction, showing how adequate design requires explicit commitments about the temporal and spatial scope of an investigation and about which level of social reality it targets. Chapter 10 treats research design as the work of aligning inquiry with the architecture of social reality, covering the movement from topic to research problem, the three pathways of ontologically informed research, the construction of theoretical frameworks, and the derivation of methodology and methods. Chapter 11 examines the major research methods across the social sciences and demonstrates how approaches to data collection and analysis embed ontological assumptions. Chapter 12 turns to the conduct of ontologically informed research in practice. Chapter 13 draws the analysis together, showing how ontological awareness functions as a foundation for more adequate social research and tracing the implications of the four debates for research practice, policy development and the ongoing refinement of social scientific knowledge.
The book rewards being read in order, particularly through Parts I and II, where each chapter and/or debate builds on the one before it. Readers who want the practical payoff before the philosophical groundwork can begin with Chapter 4, which introduces the situating practice, and return to Chapters 1 to 3 once the application case shows the necessity for the underlying questions that are covered in the first few chapters.
What the project is finally after is a way past the standing difficulty of interdisciplinarity. Interdisciplinarity is usually treated as a matter of assembling researchers from different departments or choosing topics broad enough to require them, which makes it an administrative arrangement more often than an intellectual one. What has been missing is a common language at the level where the disagreements actually lie. That is what this book tries to offer: a vocabulary that can help researchers say where they stand, understand where others stand, and get on with the work of finding out how social reality operates.
References
Allison, H. E. (2004). Kant’s transcendental idealism: An interpretation and defense (Revised ed.). Yale University Press.
Althusser, L. (1970). For Marx (B. Brewster, Trans.). Vintage Books. (Original work published 1965).
Archer, M. S. (1995). Realist social theory: The morphogenetic approach. Cambridge University Press.
Archer, M. S., Bhaskar, R., Collier, A., Lawson, T., & Norrie, A. (Eds.). (1998). Critical realism: Essential readings. Routledge.
Ayers, M. (1991). Locke: Epistemology and ontology (2 vols.). Routledge.
Barry, A., & Born, G. (Eds.). (2013). Interdisciplinarity: Reconfigurations of the social and natural sciences. Routledge.
Bhaskar, R. (1975). A realist theory of science. Leeds Books.
Bhaskar, R. (1979). The possibility of naturalism: A philosophical critique of the contemporary human sciences. Harvester Press.
Bryman, A. (1988). Quantity and quality in social research. Unwin Hyman.
Danermark, B., Ekström, M., Jakobsen, L., & Karlsson, J. C. (2002). Explaining society: Critical realism in the social sciences. Routledge.
Fogelin, R. J. (1985). Hume’s scepticism in the Treatise of Human Nature. Routledge.
Gardner, S. (1999). Kant and the Critique of Pure Reason. Routledge.
Garrett, D. (1997). Cognition and commitment in Hume’s philosophy. Oxford University Press.
Giddens, A. (1984). The constitution of society: Outline of the theory of structuration. University of California Press.
Goertz, G., & Mahoney, J. (2012). A tale of two cultures: Qualitative and quantitative research in the social sciences. Princeton University Press.
Guyer, P. (Ed.). (1992). The Cambridge companion to Kant. Cambridge University Press.
Hume, D. (1739-1740). A treatise of human nature (L. A. Selby-Bigge & P. H. Nidditch, Eds.). Oxford University Press.
Hume, D. (1748). An enquiry concerning human understanding (T. L. Beauchamp, Ed.). Oxford University Press.
Kant, I. (1781). Critique of pure reason (P. Guyer & A. W. Wood, Trans.). Cambridge University Press.
Kant, I. (1788). Critique of practical reason (A. Reath, Ed.; M. Gregor, Trans.). Cambridge University Press.
Kant, I. (1790). Critique of the power of judgment (P. Guyer & E. Matthews, Trans.). Cambridge University Press.
Kuhn, T. S. (1962). The structure of scientific revolutions. University of Chicago Press.
Lakatos, I. (1978). The methodology of scientific research programmes: Philosophical papers volume 1. Cambridge University Press.
Laudan, L. (1977). Progress and its problems: Towards a theory of scientific growth. University of California Press.
Lawson, T. (1997). Economics and reality. Routledge.
Locke, J. (1689). An essay concerning human understanding (P. H. Nidditch, Ed.). Oxford University Press.
Massey, D. (2018). Power-geometry and a progressive sense of place. In The Doreen Massey Reader (pp. 149–158). Agenda Publishing. https://doi.org/10.1017/9781911116844.011 (Original work published 1993)
Rovelli, C. (2004). Quantum gravity. Cambridge University Press.
Sayer, A. (2000). Realism and social science. Sage.
Spinoza, B. (1994). A Spinoza reader: The Ethics and other works (E. Curley, Ed. & Trans.). Princeton University Press. (Original work published 1677).
Stroud, B. (1977). Hume. Routledge.
Winch, P. (1958). The idea of a social science and its relation to philosophy. Routledge & Kegan Paul.
Footnotes
-
We use the word adequate in a particular sense throughout this book, borrowed from Spinoza (1677/1994). An account is adequate when it grasps its object through what produces it, rather than through the partial way that object happens to present itself to us. Its opposite is not falsehood but incompleteness, a part mistaken for the whole. Adequacy is therefore always adequacy to something, and it comes in degrees, which is why we describe explanations and research designs as more or less adequate to what they investigate rather than as true or false. The glossary gives the fuller definition. ↩