justification of hypothesis meaning

• Table of Contents
• Random Entry
• Chronological
• Editorial Information
• About the SEP
• Editorial Board
• How to Cite the SEP
• Special Characters
• Advanced Tools
• Support the SEP
• PDFs for SEP Friends
• Make a Donation
• SEPIA for Libraries
• Entry Contents

Bibliography

Academic tools.

• Friends PDF Preview
• Author and Citation Info
• Back to Top

Scientific Discovery

Scientific discovery is the process or product of successful scientific inquiry. Objects of discovery can be things, events, processes, causes, and properties as well as theories and hypotheses and their features (their explanatory power, for example). Most philosophical discussions of scientific discoveries focus on the generation of new hypotheses that fit or explain given data sets or allow for the derivation of testable consequences. Philosophical discussions of scientific discovery have been intricate and complex because the term “discovery” has been used in many different ways, both to refer to the outcome and to the procedure of inquiry. In the narrowest sense, the term “discovery” refers to the purported “eureka moment” of having a new insight. In the broadest sense, “discovery” is a synonym for “successful scientific endeavor” tout court. Some philosophical disputes about the nature of scientific discovery reflect these terminological variations.

Philosophical issues related to scientific discovery arise about the nature of human creativity, specifically about whether the “eureka moment” can be analyzed and about whether there are rules (algorithms, guidelines, or heuristics) according to which such a novel insight can be brought about. Philosophical issues also arise about the analysis and evaluation of heuristics, about the characteristics of hypotheses worthy of articulation and testing, and, on the meta-level, about the nature and scope of philosophical analysis itself. This essay describes the emergence and development of the philosophical problem of scientific discovery and surveys different philosophical approaches to understanding scientific discovery. In doing so, it also illuminates the meta-philosophical problems surrounding the debates, and, incidentally, the changing nature of philosophy of science.

1. Introduction

2. scientific inquiry as discovery, 3. elements of discovery, 4. pragmatic logics of discovery, 5. the distinction between the context of discovery and the context of justification, 6.1 discovery as abduction, 6.2 heuristic programming, 7. anomalies and the structure of discovery, 8.1 discoverability, 8.2 preliminary appraisal, 8.3 heuristic strategies, 9.1 kinds and features of creativity, 9.2 analogy, 9.3 mental models, 10. machine discovery, 11. social epistemology and discovery, 12. integrated approaches to knowledge generation, other internet resources, related entries.

Philosophical reflection on scientific discovery occurred in different phases. Prior to the 1930s, philosophers were mostly concerned with discoveries in the broad sense of the term, that is, with the analysis of successful scientific inquiry as a whole. Philosophical discussions focused on the question of whether there were any discernible patterns in the production of new knowledge. Because the concept of discovery did not have a specified meaning and was used in a very wide sense, almost all discussions of scientific method and practice could potentially be considered as early contributions to reflections on scientific discovery. In the course of the 18 th century, as philosophy of science and science gradually became two distinct endeavors with different audiences, the term “discovery” became a technical term in philosophical discussions. Different elements of scientific inquiry were specified. Most importantly, during the 19 th century, the generation of new knowledge came to be clearly and explicitly distinguished from its assessment, and thus the conditions for the narrower notion of discovery as the act or process of conceiving new ideas emerged. This distinction was encapsulated in the so-called “context distinction,” between the “context of discovery” and the “context of justification”.

Much of the discussion about scientific discovery in the 20 th century revolved around this distinction It was argued that conceiving a new idea is a non-rational process, a leap of insight that cannot be captured in specific instructions. Justification, by contrast, is a systematic process of applying evaluative criteria to knowledge claims. Advocates of the context distinction argued that philosophy of science is exclusively concerned with the context of justification. The assumption underlying this argument is that philosophy is a normative project; it determines norms for scientific practice. Given this assumption, only the justification of ideas, not their generation, can be the subject of philosophical (normative) analysis. Discovery, by contrast, can only be a topic for empirical study. By definition, the study of discovery is outside the scope of philosophy of science proper.

The introduction of the context distinction and the disciplinary distinction between empirical science studies and normative philosophy of science that was tied to it spawned meta-philosophical disputes. For a long time, philosophical debates about discovery were shaped by the notion that philosophical and empirical analyses are mutually exclusive. Some philosophers insisted, like their predecessors prior to the 1930s, that the philosopher’s tasks include the analysis of actual scientific practices and that scientific resources be used to address philosophical problems. They maintained that it is a legitimate task for philosophy of science to develop a theory of heuristics or problem solving. But this position was the minority view in philosophy of science until the last decades of the 20 th century. Philosophers of discovery were thus compelled to demonstrate that scientific discovery was in fact a legitimate part of philosophy of science. Philosophical reflections about the nature of scientific discovery had to be bolstered by meta-philosophical arguments about the nature and scope of philosophy of science.

Today, however, there is wide agreement that philosophy and empirical research are not mutually exclusive. Not only do empirical studies of actual scientific discoveries in past and present inform philosophical thought about the structure and cognitive mechanisms of discovery, but works in psychology, cognitive science, artificial intelligence and related fields have become integral parts of philosophical analyses of the processes and conditions of the generation of new knowledge. Social epistemology has opened up another perspective on scientific discovery, reconceptualizing knowledge generation as group process.

Prior to the 19 th century, the term “discovery” was used broadly to refer to a new finding, such as a new cure, an unknown territory, an improvement of an instrument, or a new method of measuring longitude. One strand of the discussion about discovery dating back to ancient times concerns the method of analysis as the method of discovery in mathematics and geometry, and, by extension, in philosophy and scientific inquiry. Following the analytic method, we seek to find or discover something – the “thing sought,” which could be a theorem, a solution to a geometrical problem, or a cause – by analyzing it. In the ancient Greek context, analytic methods in mathematics, geometry, and philosophy were not clearly separated; the notion of finding or discovering things by analysis was relevant in all these fields.

In the ensuing centuries, several natural and experimental philosophers, including Avicenna and Zabarella, Bacon and Boyle, the authors of the Port-Royal Logic and Newton, and many others, expounded rules of reasoning and methods for arriving at new knowledge. The ancient notion of analysis still informed these rules and methods. Newton’s famous thirty-first query in the second edition of the Opticks outlines the role of analysis in discovery as follows: “As in Mathematicks, so in Natural Philosophy, the Investigation of difficult Things by the Method of Analysis, ought ever to precede the Method of Composition. This Analysis consists in making Experiments and Observations, and in drawing general Conclusions from them by Induction, and admitting of no Objections against the Conclusions, but such as are taken from Experiments, or other certain Truths … By this way of Analysis we may proceed from Compounds to Ingredients, and from Motions to the Forces producing them; and in general, from Effects to their Causes, and from particular Causes to more general ones, till the Argument end in the most general. This is the Method of Analysis” (Newton 1718, 380, see Koertge 1980, section VI). Early modern accounts of discovery captured knowledge-seeking practices in the study of living and non-living nature, ranging from astronomy and physics to medicine, chemistry, and agriculture. These rich accounts of scientific inquiry were often expounded to bolster particular theories about the nature of matter and natural forces and were not explicitly labeled “methods of discovery ”, yet they are, in fact, accounts of knowledge generation and proper scientific reasoning, covering topics such as the role of the senses in knowledge generation, observation and experimentation, analysis and synthesis, induction and deduction, hypotheses, probability, and certainty.

Bacon’s work is a prominent example. His view of the method of science as it is presented in the Novum Organum showed how best to arrive at knowledge about “form natures” (the most general properties of matter) via a systematic investigation of phenomenal natures. Bacon described how first to collect and organize natural phenomena and experimentally produced facts in tables, how to evaluate these lists, and how to refine the initial results with the help of further trials. Through these steps, the investigator would arrive at conclusions about the “form nature” that produces particular phenomenal natures. Bacon expounded the procedures of constructing and evaluating tables of presences and absences to underpin his matter theory. In addition, in his other writings, such as his natural history Sylva Sylvarum or his comprehensive work on human learning De Augmentis Scientiarium , Bacon exemplified the “art of discovery” with practical examples and discussions of strategies of inquiry.

Like Bacon and Newton, several other early modern authors advanced ideas about how to generate and secure empirical knowledge, what difficulties may arise in scientific inquiry, and how they could be overcome. The close connection between theories about matter and force and scientific methodologies that we find in early modern works was gradually severed. 18 th - and early 19 th -century authors on scientific method and logic cited early modern approaches mostly to model proper scientific practice and reasoning, often creatively modifying them ( section 3 ). Moreover, they developed the earlier methodologies of experimentation, observation, and reasoning into practical guidelines for discovering new phenomena and devising probable hypotheses about cause-effect relations.

It was common in 20 th -century philosophy of science to draw a sharp contrast between those early theories of scientific method and modern approaches. 20 th -century philosophers of science interpreted 17 th - and 18 th -century approaches as generative theories of scientific method. They function simultaneously as guides for acquiring new knowledge and as assessments of the knowledge thus obtained, whereby knowledge that is obtained “in the right way” is considered secure (Laudan 1980; Schaffner 1993: chapter 2). On this view, scientific methods are taken to have probative force (Nickles 1985). According to modern, “consequentialist” theories, propositions must be established by comparing their consequences with observed and experimentally produced phenomena (Laudan 1980; Nickles 1985). It was further argued that, when consequentialist theories were on the rise, the two processes of generation and assessment of an idea or hypothesis became distinct, and the view that the merit of a new idea does not depend on the way in which it was arrived at became widely accepted.

More recent research in history of philosophy of science has shown, however, that there was no such sharp contrast. Consequentialist ideas were advanced throughout the 18 th century, and the early modern generative theories of scientific method and knowledge were more pragmatic than previously assumed. Early modern scholars did not assume that this procedure would lead to absolute certainty. One could only obtain moral certainty for the propositions thus secured.

During the 18 th and 19 th centuries, the different elements of discovery gradually became separated and discussed in more detail. Discussions concerned the nature of observations and experiments, the act of having an insight and the processes of articulating, developing, and testing the novel insight. Philosophical discussion focused on the question of whether and to what extent rules could be devised to guide each of these processes.

Numerous 19 th -century scholars contributed to these discussions, including Claude Bernard, Auguste Comte, George Gore, John Herschel, W. Stanley Jevons, Justus von Liebig, John Stuart Mill, and Charles Sanders Peirce, to name only a few. William Whewell’s work, especially the two volumes of Philosophy of the Inductive Sciences of 1840, is a noteworthy and, later, much discussed contribution to the philosophical debates about scientific discovery because he explicitly distinguished the creative moment or “happy thought” as he called it from other elements of scientific inquiry and because he offered a detailed analysis of the “discoverer’s induction”, i.e., the pursuit and evaluation of the new insight. Whewell’s approach is not unique, but for late 20 th -century philosophers of science, his comprehensive, historically informed philosophy of discovery became a point of orientation in the revival of interest in scientific discovery processes.

For Whewell, discovery comprised three elements: the happy thought, the articulation and development of that thought, and the testing or verification of it. His account was in part a description of the psychological makeup of the discoverer. For instance, he held that only geniuses could have those happy thoughts that are essential to discovery. In part, his account was an account of the methods by which happy thoughts are integrated into the system of knowledge. According to Whewell, the initial step in every discovery is what he called “some happy thought, of which we cannot trace the origin; some fortunate cast of intellect, rising above all rules. No maxims can be given which inevitably lead to discovery” (Whewell 1996 [1840]: 186). An “art of discovery” in the sense of a teachable and learnable skill does not exist according to Whewell. The happy thought builds on the known facts, but according to Whewell it is impossible to prescribe a method for having happy thoughts.

In this sense, happy thoughts are accidental. But in an important sense, scientific discoveries are not accidental. The happy thought is not a wild guess. Only the person whose mind is prepared to see things will actually notice them. The “previous condition of the intellect, and not the single fact, is really the main and peculiar cause of the success. The fact is merely the occasion by which the engine of discovery is brought into play sooner or later. It is, as I have elsewhere said, only the spark which discharges a gun already loaded and pointed; and there is little propriety in speaking of such an accident as the cause why the bullet hits its mark.” (Whewell 1996 [1840]: 189).

Having a happy thought is not yet a discovery, however. The second element of a scientific discovery consists in binding together—“colligating”, as Whewell called it—a set of facts by bringing them under a general conception. Not only does the colligation produce something new, but it also shows the previously known facts in a new light. Colligation involves, on the one hand, the specification of facts through systematic observation, measurements and experiment, and on the other hand, the clarification of ideas through the exposition of the definitions and axioms that are tacitly implied in those ideas. This process is extended and iterative. The scientists go back and forth between binding together the facts, clarifying the idea, rendering the facts more exact, and so forth.

The final part of the discovery is the verification of the colligation involving the happy thought. This means, first and foremost, that the outcome of the colligation must be sufficient to explain the data at hand. Verification also involves judging the predictive power, simplicity, and “consilience” of the outcome of the colligation. “Consilience” refers to a higher range of generality (broader applicability) of the theory (the articulated and clarified happy thought) that the actual colligation produced. Whewell’s account of discovery is not a deductivist system. It is essential that the outcome of the colligation be inferable from the data prior to any testing (Snyder 1997).

Whewell’s theory of discovery clearly separates three elements: the non-analyzable happy thought or eureka moment; the process of colligation which includes the clarification and explication of facts and ideas; and the verification of the outcome of the colligation. His position that the philosophy of discovery cannot prescribe how to think happy thoughts has been a key element of 20 th -century philosophical reflection on discovery. In contrast to many 20 th -century approaches, Whewell’s philosophical conception of discovery also comprises the processes by which the happy thoughts are articulated. Similarly, the process of verification is an integral part of discovery. The procedures of articulation and test are both analyzable according to Whewell, and his conception of colligation and verification serve as guidelines for how the discoverer should proceed. To verify a hypothesis, the investigator needs to show that it accounts for the known facts, that it foretells new, previously unobserved phenomena, and that it can explain and predict phenomena which are explained and predicted by a hypothesis that was obtained through an independent happy thought-cum-colligation (Ducasse 1951).

Whewell’s conceptualization of scientific discovery offers a useful framework for mapping the philosophical debates about discovery and for identifying major issues of concern in 20 th -century philosophical debates. Until the late 20 th century, most philosophers operated with a notion of discovery that is narrower than Whewell’s. In more recent treatments of discovery, however, the scope of the term “discovery” is limited to either the first of these elements, the “happy thought”, or to the happy thought and its initial articulation. In the narrower conception, what Whewell called “verification” is not part of discovery proper. Secondly, until the late 20 th century, there was wide agreement that the eureka moment, narrowly construed, is an unanalyzable, even mysterious leap of insight. The main disagreements concerned the question of whether the process of developing a hypothesis (the “colligation” in Whewell’s terms) is, or is not, a part of discovery proper – and if it is, whether and how this process is guided by rules. Much of the controversies in the 20 th century about the possibility of a philosophy of discovery can be understood against the background of the disagreement about whether the process of discovery does or does not include the articulation and development of a novel thought. Philosophers also disagreed on the issue of whether it is a philosophical task to explicate these rules.

In early 20 th -century logical empiricism, the view that discovery is or at least crucially involves a non-analyzable creative act of a gifted genius was widespread. Alternative conceptions of discovery especially in the pragmatist tradition emphasize that discovery is an extended process, i.e., that the discovery process includes the reasoning processes through which a new insight is articulated and further developed.

In the pragmatist tradition, the term “logic” is used in the broad sense to refer to strategies of human reasoning and inquiry. While the reasoning involved does not proceed according to the principles of demonstrative logic, it is systematic enough to deserve the label “logical”. Proponents of this view argued that traditional (here: syllogistic) logic is an inadequate model of scientific discovery because it misrepresents the process of knowledge generation as grossly as the notion of an “aha moment”.

Early 20 th -century pragmatic logics of discovery can best be described as comprehensive theories of the mental and physical-practical operations involved in knowledge generation, as theories of “how we think” (Dewey 1910). Among the mental operations are classification, determination of what is relevant to an inquiry, and the conditions of communication of meaning; among the physical operations are observation and (laboratory) experiments. These features of scientific discovery are either not or only insufficiently represented by traditional syllogistic logic (Schiller 1917: 236–7).

Philosophers advocating this approach agree that the logic of discovery should be characterized as a set of heuristic principles rather than as a process of applying inductive or deductive logic to a set of propositions. These heuristic principles are not understood to show the path to secure knowledge. Heuristic principles are suggestive rather than demonstrative (Carmichael 1922, 1930). One recurrent feature in these accounts of the reasoning strategies leading to new ideas is analogical reasoning (Schiller 1917; Benjamin 1934, see also section 9.2 .). However, in academic philosophy of science, endeavors to develop more systematically the heuristics guiding discovery processes were soon eclipsed by the advance of the distinction between contexts of discovery and justification.

The distinction between “context of discovery” and “context of justification” dominated and shaped the discussions about discovery in 20 th -century philosophy of science. The context distinction marks the distinction between the generation of a new idea or hypothesis and the defense (test, verification) of it. As the previous sections have shown, the distinction among different elements of scientific inquiry has a long history but in the first half of the 20 th century, the distinction between the different features of scientific inquiry turned into a powerful demarcation criterion between “genuine” philosophy and other fields of science studies, which became potent in philosophy of science. The boundary between context of discovery (the de facto thinking processes) and context of justification (the de jure defense of the correctness of these thoughts) was now understood to determine the scope of philosophy of science, whereby philosophy of science is conceived as a normative endeavor. Advocates of the context distinction argue that the generation of a new idea is an intuitive, nonrational process; it cannot be subject to normative analysis. Therefore, the study of scientists’ actual thinking can only be the subject of psychology, sociology, and other empirical sciences. Philosophy of science, by contrast, is exclusively concerned with the context of justification.

The terms “context of discovery” and “context of justification” are often associated with Hans Reichenbach’s work. Reichenbach’s original conception of the context distinction is quite complex, however (Howard 2006; Richardson 2006). It does not map easily on to the disciplinary distinction mentioned above, because for Reichenbach, philosophy of science proper is partly descriptive. Reichenbach maintains that philosophy of science includes a description of knowledge as it really is. Descriptive philosophy of science reconstructs scientists’ thinking processes in such a way that logical analysis can be performed on them, and it thus prepares the ground for the evaluation of these thoughts (Reichenbach 1938: § 1). Discovery, by contrast, is the object of empirical—psychological, sociological—study. According to Reichenbach, the empirical study of discoveries shows that processes of discovery often correspond to the principle of induction, but this is simply a psychological fact (Reichenbach 1938: 403).

While the terms “context of discovery” and “context of justification” are widely used, there has been ample discussion about how the distinction should be drawn and what their philosophical significance is (c.f. Kordig 1978; Gutting 1980; Zahar 1983; Leplin 1987; Hoyningen-Huene 1987; Weber 2005: chapter 3; Schickore and Steinle 2006). Most commonly, the distinction is interpreted as a distinction between the process of conceiving a theory and the assessment of that theory, specifically the assessment of the theory’s epistemic support. This version of the distinction is not necessarily interpreted as a temporal distinction. In other words, it is not usually assumed that a theory is first fully developed and then assessed. Rather, generation and assessment are two different epistemic approaches to theory: the endeavor to articulate, flesh out, and develop its potential and the endeavor to assess its epistemic worth. Within the framework of the context distinction, there are two main ways of conceptualizing the process of conceiving a theory. The first option is to characterize the generation of new knowledge as an irrational act, a mysterious creative intuition, a “eureka moment”. The second option is to conceptualize the generation of new knowledge as an extended process that includes a creative act as well as some process of articulating and developing the creative idea.

Both of these accounts of knowledge generation served as starting points for arguments against the possibility of a philosophy of discovery. In line with the first option, philosophers have argued that neither is it possible to prescribe a logical method that produces new ideas nor is it possible to reconstruct logically the process of discovery. Only the process of testing is amenable to logical investigation. This objection to philosophies of discovery has been called the “discovery machine objection” (Curd 1980: 207). It is usually associated with Karl Popper’s Logic of Scientific Discovery .

The initial state, the act of conceiving or inventing a theory, seems to me neither to call for logical analysis not to be susceptible of it. The question how it happens that a new idea occurs to a man—whether it is a musical theme, a dramatic conflict, or a scientific theory—may be of great interest to empirical psychology; but it is irrelevant to the logical analysis of scientific knowledge. This latter is concerned not with questions of fact (Kant’s quid facti ?) , but only with questions of justification or validity (Kant’s quid juris ?) . Its questions are of the following kind. Can a statement be justified? And if so, how? Is it testable? Is it logically dependent on certain other statements? Or does it perhaps contradict them? […]Accordingly I shall distinguish sharply between the process of conceiving a new idea, and the methods and results of examining it logically. As to the task of the logic of knowledge—in contradistinction to the psychology of knowledge—I shall proceed on the assumption that it consists solely in investigating the methods employed in those systematic tests to which every new idea must be subjected if it is to be seriously entertained. (Popper 2002 [1934/1959]: 7–8)

With respect to the second way of conceptualizing knowledge generation, many philosophers argue in a similar fashion that because the process of discovery involves an irrational, intuitive process, which cannot be examined logically, a logic of discovery cannot be construed. Other philosophers turn against the philosophy of discovery even though they explicitly acknowledge that discovery is an extended, reasoned process. They present a meta-philosophical objection argument, arguing that a theory of articulating and developing ideas is not a philosophical but a psychological or sociological theory. In this perspective, “discovery” is understood as a retrospective label, which is attributed as a sign of accomplishment to some scientific endeavors. Sociological theories acknowledge that discovery is a collective achievement and the outcome of a process of negotiation through which “discovery stories” are constructed and certain knowledge claims are granted discovery status (Brannigan 1981; Schaffer 1986, 1994).

The impact of the context distinction on 20 th -century studies of scientific discovery and on philosophy of science more generally can hardly be overestimated. The view that the process of discovery (however construed) is outside the scope of philosophy of science proper was widely shared amongst philosophers of science for most of the 20 th century. The last section shows that there were some attempts to develop logics of discovery in the 1920s and 1930s, especially in the pragmatist tradition. But for several decades, the context distinction dictated what philosophy of science should be about and how it should proceed. The dominant view was that theories of mental operations or heuristics had no place in philosophy of science and that, therefore, discovery was not a legitimate topic for philosophy of science. Until the last decades of the 20 th century, there were few attempts to challenge the disciplinary distinction tied to the context distinction. Only during the 1970s did the interest in philosophical approaches to discovery begin to increase again. But the context distinction remained a challenge for philosophies of discovery.

There are several lines of response to the disciplinary distinction tied to the context distinction. Each of these lines of response opens a philosophical perspective on discovery. Each proceeds on the assumption that philosophy of science may legitimately include some form of analysis of actual reasoning patterns as well as information from empirical sciences such as cognitive science, psychology, and sociology. All of these responses reject the idea that discovery is nothing but a mystical event. Discovery is conceived as an analyzable reasoning process, not just as a creative leap by which novel ideas spring into being fully formed. All of these responses agree that the procedures and methods for arriving at new hypotheses and ideas are no guarantee that the hypothesis or idea that is thus formed is necessarily the best or the correct one. Nonetheless, it is the task of philosophy of science to provide rules for making this process better. All of these responses can be described as theories of problem solving, whose ultimate goal is to make the generation of new ideas and theories more efficient.

But the different approaches to scientific discovery employ different terminologies. In particular, the term “logic” of discovery is sometimes used in a narrow sense and sometimes broadly understood. In the narrow sense, “logic” of discovery is understood to refer to a set of formal, generally applicable rules by which novel ideas can be mechanically derived from existing data. In the broad sense, “logic” of discovery refers to the schematic representation of reasoning procedures. “Logical” is just another term for “rational”. Moreover, while each of these responses combines philosophical analyses of scientific discovery with empirical research on actual human cognition, different sets of resources are mobilized, ranging from AI research and cognitive science to historical studies of problem-solving procedures. Also, the responses parse the process of scientific inquiry differently. Often, scientific inquiry is regarded as having two aspects, viz. generation and assessments of new ideas. At times, however, scientific inquiry is regarded as having three aspects, namely generation, pursuit or articulation, and assessment of knowledge. In the latter framework, the label “discovery” is sometimes used to refer just to generation and sometimes to refer to both generation and pursuit.

One response to the challenge of the context distinction draws on a broad understanding of the term “logic” to argue that we cannot but admit a general, domain-neutral logic if we do not want to assume that the success of science is a miracle (Jantzen 2016) and that a logic of scientific discovery can be developed ( section 6 ). Another response, drawing on a narrow understanding of the term “logic”, is to concede that there is no logic of discovery, i.e., no algorithm for generating new knowledge, but that the process of discovery follows an identifiable, analyzable pattern ( section 7 ).

Others argue that discovery is governed by a methodology . The methodology of discovery is a legitimate topic for philosophical analysis ( section 8 ). Yet another response assumes that discovery is or at least involves a creative act. Drawing on resources from cognitive science, neuroscience, computational research, and environmental and social psychology, philosophers have sought to demystify the cognitive processes involved in the generation of new ideas. Philosophers who take this approach argue that scientific creativity is amenable to philosophical analysis ( section 9.1 ).

All these responses assume that there is more to discovery than a eureka moment. Discovery comprises processes of articulating, developing, and assessing the creative thought, as well as the scientific community’s adjudication of what does, and does not, count as “discovery” (Arabatzis 1996). These are the processes that can be examined with the tools of philosophical analysis, augmented by input from other fields of science studies such as sociology, history, or cognitive science.

6. Logics of discovery after the context distinction

One way of responding to the demarcation criterion described above is to argue that discovery is a topic for philosophy of science because it is a logical process after all. Advocates of this approach to the logic of discovery usually accept the overall distinction between the two processes of conceiving and testing a hypothesis. They also agree that it is impossible to put together a manual that provides a formal, mechanical procedure through which innovative concepts or hypotheses can be derived: There is no discovery machine. But they reject the view that the process of conceiving a theory is a creative act, a mysterious guess, a hunch, a more or less instantaneous and random process. Instead, they insist that both conceiving and testing hypotheses are processes of reasoning and systematic inference, that both of these processes can be represented schematically, and that it is possible to distinguish better and worse paths to new knowledge.

This line of argument has much in common with the logics of discovery described in section 4 above but it is now explicitly pitched against the disciplinary distinction tied to the context distinction. There are two main ways of developing this argument. The first is to conceive of discovery in terms of abductive reasoning ( section 6.1 ). The second is to conceive of discovery in terms of problem-solving algorithms, whereby heuristic rules aid the processing of available data and enhance the success in finding solutions to problems ( section 6.2 ). Both lines of argument rely on a broad conception of logic, whereby the “logic” of discovery amounts to a schematic account of the reasoning processes involved in knowledge generation.

One argument, elaborated prominently by Norwood R. Hanson, is that the act of discovery—here, the act of suggesting a new hypothesis—follows a distinctive logical pattern, which is different from both inductive logic and the logic of hypothetico-deductive reasoning. The special logic of discovery is the logic of abductive or “retroductive” inferences (Hanson 1958). The argument that it is through an act of abductive inferences that plausible, promising scientific hypotheses are devised goes back to C.S. Peirce. This version of the logic of discovery characterizes reasoning processes that take place before a new hypothesis is ultimately justified. The abductive mode of reasoning that leads to plausible hypotheses is conceptualized as an inference beginning with data or, more specifically, with surprising or anomalous phenomena.

In this view, discovery is primarily a process of explaining anomalies or surprising, astonishing phenomena. The scientists’ reasoning proceeds abductively from an anomaly to an explanatory hypothesis in light of which the phenomena would no longer be surprising or anomalous. The outcome of this reasoning process is not one single specific hypothesis but the delineation of a type of hypotheses that is worthy of further attention (Hanson 1965: 64). According to Hanson, the abductive argument has the following schematic form (Hanson 1960: 104):

• Some surprising, astonishing phenomena p 1 , p 2 , p 3 … are encountered.
• But p 1 , p 2 , p 3 … would not be surprising were an hypothesis of H ’s type to obtain. They would follow as a matter of course from something like H and would be explained by it.
• Therefore there is good reason for elaborating an hypothesis of type H—for proposing it as a possible hypothesis from whose assumption p 1 , p 2 , p 3 … might be explained.

Drawing on the historical record, Hanson argues that several important discoveries were made relying on abductive reasoning, such as Kepler’s discovery of the elliptic orbit of Mars (Hanson 1958). It is now widely agreed, however, that Hanson’s reconstruction of the episode is not a historically adequate account of Kepler’s discovery (Lugg 1985). More importantly, while there is general agreement that abductive inferences are frequent in both everyday and scientific reasoning, these inferences are no longer considered as logical inferences. Even if one accepts Hanson’s schematic representation of the process of identifying plausible hypotheses, this process is a “logical” process only in the widest sense whereby the term “logical” is understood as synonymous with “rational”. Notably, some philosophers have even questioned the rationality of abductive inferences (Koehler 1991; Brem and Rips 2000).

Another argument against the above schema is that it is too permissive. There will be several hypotheses that are explanations for phenomena p 1 , p 2 , p 3 …, so the fact that a particular hypothesis explains the phenomena is not a decisive criterion for developing that hypothesis (Harman 1965; see also Blackwell 1969). Additional criteria are required to evaluate the hypothesis yielded by abductive inferences.

Finally, it is worth noting that the schema of abductive reasoning does not explain the very act of conceiving a hypothesis or hypothesis-type. The processes by which a new idea is first articulated remain unanalyzed in the above schema. The schema focuses on the reasoning processes by which an exploratory hypothesis is assessed in terms of its merits and promise (Laudan 1980; Schaffner 1993).

In more recent work on abduction and discovery, two notions of abduction are sometimes distinguished: the common notion of abduction as inference to the best explanation (selective abduction) and creative abduction (Magnani 2000, 2009). Selective abduction—the inference to the best explanation—involves selecting a hypothesis from a set of known hypotheses. Medical diagnosis exemplifies this kind of abduction. Creative abduction, by contrast, involves generating a new, plausible hypothesis. This happens, for instance, in medical research, when the notion of a new disease is articulated. However, it is still an open question whether this distinction can be drawn, or whether there is a more gradual transition from selecting an explanatory hypothesis from a familiar domain (selective abduction) to selecting a hypothesis that is slightly modified from the familiar set and to identifying a more drastically modified or altered assumption.

Another recent suggestion is to broaden Peirce’s original account of abduction and to include not only verbal information but also non-verbal mental representations, such as visual, auditory, or motor representations. In Thagard’s approach, representations are characterized as patterns of activity in mental populations (see also section 9.3 below). The advantage of the neural account of human reasoning is that it covers features such as the surprise that accompanies the generation of new insights or the visual and auditory representations that contribute to it. Surprise, for instance, could be characterized as resulting from rapid changes in activation of the node in a neural network representing the “surprising” element (Thagard and Stewart 2011). If all mental representations can be characterized as patterns of firing in neural populations, abduction can be analyzed as the combination or “convolution” (Thagard) of patterns of neural activity from disjoint or overlapping patterns of activity (Thagard 2010).

The concern with the logic of discovery has also motivated research on artificial intelligence at the intersection of philosophy of science and cognitive science. In this approach, scientific discovery is treated as a form of problem-solving activity (Simon 1973; see also Newell and Simon 1971), whereby the systematic aspects of problem solving are studied within an information-processing framework. The aim is to clarify with the help of computational tools the nature of the methods used to discover scientific hypotheses. These hypotheses are regarded as solutions to problems. Philosophers working in this tradition build computer programs employing methods of heuristic selective search (e.g., Langley et al. 1987). In computational heuristics, search programs can be described as searches for solutions in a so-called “problem space” in a certain domain. The problem space comprises all possible configurations in that domain (e.g., for chess problems, all possible arrangements of pieces on a board of chess). Each configuration is a “state” of the problem space. There are two special states, namely the goal state, i.e., the state to be reached, and the initial state, i.e., the configuration at the starting point from which the search begins. There are operators, which determine the moves that generate new states from the current state. There are path constraints, which limit the permitted moves. Problem solving is the process of searching for a solution of the problem of how to generate the goal state from an initial state. In principle, all states can be generated by applying the operators to the initial state, then to the resulting state, until the goal state is reached (Langley et al. 1987: chapter 9). A problem solution is a sequence of operations leading from the initial to the goal state.

The basic idea behind computational heuristics is that rules can be identified that serve as guidelines for finding a solution to a given problem quickly and efficiently by avoiding undesired states of the problem space. These rules are best described as rules of thumb. The aim of constructing a logic of discovery thus becomes the aim of constructing a heuristics for the efficient search for solutions to problems. The term “heuristic search” indicates that in contrast to algorithms, problem-solving procedures lead to results that are merely provisional and plausible. A solution is not guaranteed, but heuristic searches are advantageous because they are more efficient than exhaustive random trial and error searches. Insofar as it is possible to evaluate whether one set of heuristics is better—more efficacious—than another, the logic of discovery turns into a normative theory of discovery.

Arguably, because it is possible to reconstruct important scientific discovery processes with sets of computational heuristics, the scientific discovery process can be considered as a special case of the general mechanism of information processing. In this context, the term “logic” is not used in the narrow sense of a set of formal, generally applicable rules to draw inferences but again in a broad sense as a label for a set of procedural rules.

The computer programs that embody the principles of heuristic searches in scientific inquiry simulate the paths that scientists followed when they searched for new theoretical hypotheses. Computer programs such as BACON (Simon et al. 1981) and KEKADA (Kulkarni and Simon 1988) utilize sets of problem-solving heuristics to detect regularities in given data sets. The program would note, for instance, that the values of a dependent term are constant or that a set of values for a term x and a set of values for a term y are linearly related. It would thus “infer” that the dependent term always has that value or that a linear relation exists between x and y . These programs can “make discoveries” in the sense that they can simulate successful discoveries such as Kepler’s third law (BACON) or the Krebs cycle (KEKADA).

Computational theories of scientific discoveries have helped identify and clarify a number of problem-solving strategies. An example of such a strategy is heuristic means-ends analysis, which involves identifying specific differences between the present and the goal situation and searches for operators (processes that will change the situation) that are associated with the differences that were detected. Another important heuristic is to divide the problem into sub-problems and to begin solving the one with the smallest number of unknowns to be determined (Simon 1977). Computational approaches have also highlighted the extent to which the generation of new knowledge draws on existing knowledge that constrains the development of new hypotheses.

As accounts of scientific discoveries, the early computational heuristics have some limitations. Compared to the problem spaces given in computational heuristics, the complex problem spaces for scientific problems are often ill defined, and the relevant search space and goal state must be delineated before heuristic assumptions could be formulated (Bechtel and Richardson 1993: chapter 1). Because a computer program requires the data from actual experiments, the simulations cover only certain aspects of scientific discoveries; in particular, it cannot determine by itself which data is relevant, which data to relate and what form of law it should look for (Gillies 1996). However, as a consequence of the rise of so-called “deep learning” methods in data-intensive science, there is renewed philosophical interest in the question of whether machines can make discoveries ( section 10 ).

Many philosophers maintain that discovery is a legitimate topic for philosophy of science while abandoning the notion that there is a logic of discovery. One very influential approach is Thomas Kuhn’s analysis of the emergence of novel facts and theories (Kuhn 1970 [1962]: chapter 6). Kuhn identifies a general pattern of discovery as part of his account of scientific change. A discovery is not a simple act, but an extended, complex process, which culminates in paradigm changes. Paradigms are the symbolic generalizations, metaphysical commitments, values, and exemplars that are shared by a community of scientists and that guide the research of that community. Paradigm-based, normal science does not aim at novelty but instead at the development, extension, and articulation of accepted paradigms. A discovery begins with an anomaly, that is, with the recognition that the expectations induced by an established paradigm are being violated. The process of discovery involves several aspects: observations of an anomalous phenomenon, attempts to conceptualize it, and changes in the paradigm so that the anomaly can be accommodated.

It is the mark of success of normal science that it does not make transformative discoveries, and yet such discoveries come about as a consequence of normal, paradigm-guided science. The more detailed and the better developed a paradigm, the more precise are its predictions. The more precisely the researchers know what to expect, the better they are able to recognize anomalous results and violations of expectations:

novelty ordinarily emerges only for the man who, knowing with precision what he should expect, is able to recognize that something has gone wrong. Anomaly appears only against the background provided by the paradigm. (Kuhn 1970 [1962]: 65)

Drawing on several historical examples, Kuhn argues that it is usually impossible to identify the very moment when something was discovered or even the individual who made the discovery. Kuhn illustrates these points with the discovery of oxygen (see Kuhn 1970 [1962]: 53–56). Oxygen had not been discovered before 1774 and had been discovered by 1777. Even before 1774, Lavoisier had noticed that something was wrong with phlogiston theory, but he was unable to move forward. Two other investigators, C. W. Scheele and Joseph Priestley, independently identified a gas obtained from heating solid substances. But Scheele’s work remained unpublished until after 1777, and Priestley did not identify his substance as a new sort of gas. In 1777, Lavoisier presented the oxygen theory of combustion, which gave rise to fundamental reconceptualization of chemistry. But according to this theory as Lavoisier first presented it, oxygen was not a chemical element. It was an atomic “principle of acidity” and oxygen gas was a combination of that principle with caloric. According to Kuhn, all of these developments are part of the discovery of oxygen, but none of them can be singled out as “the” act of discovery.

In pre-paradigmatic periods or in times of paradigm crisis, theory-induced discoveries may happen. In these periods, scientists speculate and develop tentative theories, which may lead to novel expectations and experiments and observations to test whether these expectations can be confirmed. Even though no precise predictions can be made, phenomena that are thus uncovered are often not quite what had been expected. In these situations, the simultaneous exploration of the new phenomena and articulation of the tentative hypotheses together bring about discovery.

In cases like the discovery of oxygen, by contrast, which took place while a paradigm was already in place, the unexpected becomes apparent only slowly, with difficulty, and against some resistance. Only gradually do the anomalies become visible as such. It takes time for the investigators to recognize “both that something is and what it is” (Kuhn 1970 [1962]: 55). Eventually, a new paradigm becomes established and the anomalous phenomena become the expected phenomena.

Recent studies in cognitive neuroscience of brain activity during periods of conceptual change support Kuhn’s view that conceptual change is hard to achieve. These studies examine the neural processes that are involved in the recognition of anomalies and compare them with the brain activity involved in the processing of information that is consistent with preferred theories. The studies suggest that the two types of data are processed differently (Dunbar et al. 2007).

8. Methodologies of discovery

Advocates of the view that there are methodologies of discovery use the term “logic” in the narrow sense of an algorithmic procedure to generate new ideas. But like the AI-based theories of scientific discovery described in section 6 , methodologies of scientific discovery interpret the concept “discovery” as a label for an extended process of generating and articulating new ideas and often describe the process in terms of problem solving. In these approaches, the distinction between the contexts of discovery and the context of justification is challenged because the methodology of discovery is understood to play a justificatory role. Advocates of a methodology of discovery usually rely on a distinction between different justification procedures, justification involved in the process of generating new knowledge and justification involved in testing it. Consequential or “strong” justifications are methods of testing. The justification involved in discovery, by contrast, is conceived as generative (as opposed to consequential) justification ( section 8.1 ) or as weak (as opposed to strong) justification ( section 8.2 ). Again, some terminological ambiguity exists because according to some philosophers, there are three contexts, not two: Only the initial conception of a new idea (the creative act is the context of discovery proper, and between it and justification there exists a separate context of pursuit (Laudan 1980). But many advocates of methodologies of discovery regard the context of pursuit as an integral part of the process of justification. They retain the notion of two contexts and re-draw the boundaries between the contexts of discovery and justification as they were drawn in the early 20 th century.

The methodology of discovery has sometimes been characterized as a form of justification that is complementary to the methodology of testing (Nickles 1984, 1985, 1989). According to the methodology of testing, empirical support for a theory results from successfully testing the predictive consequences derived from that theory (and appropriate auxiliary assumptions). In light of this methodology, justification for a theory is “consequential justification,” the notion that a hypothesis is established if successful novel predictions are derived from the theory or claim. Generative justification complements consequential justification. Advocates of generative justification hold that there exists an important form of justification in science that involves reasoning to a claim from data or previously established results more generally.

One classic example for a generative methodology is the set of Newton’s rules for the study of natural philosophy. According to these rules, general propositions are established by deducing them from the phenomena. The notion of generative justification seeks to preserve the intuition behind classic conceptions of justification by deduction. Generative justification amounts to the rational reconstruction of the discovery path in order to establish its discoverability had the researchers known what is known now, regardless of how it was first thought of (Nickles 1985, 1989). The reconstruction demonstrates in hindsight that the claim could have been discovered in this manner had the necessary information and techniques been available. In other words, generative justification—justification as “discoverability” or “potential discovery”—justifies a knowledge claim by deriving it from results that are already established. While generative justification does not retrace exactly those steps of the actual discovery path that were actually taken, it is a better representation of scientists’ actual practices than consequential justification because scientists tend to construe new claims from available knowledge. Generative justification is a weaker version of the traditional ideal of justification by deduction from the phenomena. Justification by deduction from the phenomena is complete if a theory or claim is completely determined from what we already know. The demonstration of discoverability results from the successful derivation of a claim or theory from the most basic and most solidly established empirical information.

Discoverability as described in the previous paragraphs is a mode of justification. Like the testing of novel predictions derived from a hypothesis, generative justification begins when the phase of finding and articulating a hypothesis worthy of assessing is drawing to a close. Other approaches to the methodology of discovery are directly concerned with the procedures involved in devising new hypotheses. The argument in favor of this kind of methodology is that the procedures of devising new hypotheses already include elements of appraisal. These preliminary assessments have been termed “weak” evaluation procedures (Schaffner 1993). Weak evaluations are relevant during the process of devising a new hypothesis. They provide reasons for accepting a hypothesis as promising and worthy of further attention. Strong evaluations, by contrast, provide reasons for accepting a hypothesis as (approximately) true or confirmed. Both “generative” and “consequential” testing as discussed in the previous section are strong evaluation procedures. Strong evaluation procedures are rigorous and systematically organized according to the principles of hypothesis derivation or H-D testing. A methodology of preliminary appraisal, by contrast, articulates criteria for the evaluation of a hypothesis prior to rigorous derivation or testing. It aids the decision about whether to take that hypothesis seriously enough to develop it further and test it. For advocates of this version of the methodology of discovery, it is the task of philosophy of science to characterize sets of constraints and methodological rules guiding the complex process of prior-to-test evaluation of hypotheses.

In contrast to the computational approaches discussed above, strategies of preliminary appraisal are not regarded as subject-neutral but as specific to particular fields of study. Philosophers of biology, for instance, have developed a fine-grained framework to account for the generation and preliminary evaluation of biological mechanisms (Darden 2002; Craver 2002; Bechtel and Richardson 1993; Craver and Darden 2013). Some philosophers have suggested that the phase of preliminary appraisal be further divided into two phases, the phase of appraising and the phase of revising. According to Lindley Darden, the phases of generation, appraisal and revision of descriptions of mechanisms can be characterized as reasoning processes governed by reasoning strategies. Different reasoning strategies govern the different phases (Darden 1991, 2002; Craver 2002; Darden 2009). The generation of hypotheses about mechanisms, for instance, is governed by the strategy of “schema instantiation” (see Darden 2002). The discovery of the mechanism of protein synthesis involved the instantiation of an abstract schema for chemical reactions: reactant 1 + reactant 2 = product. The actual mechanism of protein synthesis was found through specification and modification of this schema.

Neither of these strategies is deemed necessary for discovery, and they are not prescriptions for biological research. Rather, these strategies are deemed sufficient for the discovery of mechanisms. The methodology of the discovery of mechanisms is an extrapolation from past episodes of research on mechanisms and the result of a synthesis of rational reconstructions of several of these historical episodes. The methodology of discovery is weakly normative in the sense that the strategies for the discovery of mechanisms that were successful in the past may prove useful in future biological research (Darden 2002).

As philosophers of science have again become more attuned to actual scientific practices, interest in heuristic strategies has also been revived. Many analysts now agree that discovery processes can be regarded as problem solving activities, whereby a discovery is a solution to a problem. Heuristics-based methodologies of discovery are neither purely subjective and intuitive nor algorithmic or formalizable; the point is that reasons can be given for pursuing one or the other problem-solving strategy. These rules are open and do not guarantee a solution to a problem when applied (Ippoliti 2018). On this view, scientific researchers are no longer seen as Kuhnian “puzzle solvers” but as problem solvers and decision makers in complex, variable, and changing environments (Wimsatt 2007).

Philosophers of discovery working in this tradition draw on a growing body of literature in cognitive psychology, management science, operations research, and economy on human reasoning and decision making in contexts with limited information, under time constraints, and with sub-optimal means (Gigerenzer & Sturm 2012). Heuristic strategies characterized in these studies, such as Gigerenzer’s “tools to theory heuristic” are then applied to understand scientific knowledge generation (Gigerenzer 1992, Nickles 2018). Other analysts specify heuristic strategies in a range of scientific fields, including climate science, neurobiology, and clinical medicine (Gramelsberger 2011, Schaffner 2008, Gillies 2018). Finally, in analytic epistemology, formal methods are developed to identify and assess distinct heuristic strategies currently in use, such as Bayesian reverse engineering in cognitive science (Zednik and Jäkel 2016).

As the literature on heuristics continues to grow, it has become clear that the term “heuristics” is itself used in a variety of different ways. (For a valuable taxonomy of meanings of “heuristic,” see Chow 2015, see also Ippoliti 2018.) Moreover, as in the context of earlier debates about computational heuristics, debates continue about the limitations of heuristics. The use of heuristics may come at a cost if heuristics introduce systematic biases (Wimsatt 2007). Some philosophers thus call for general principles for the evaluation of heuristic strategies (Hey 2016).

9. Cognitive perspectives on discovery

The approaches to scientific discovery presented in the previous sections focus on the adoption, articulation, and preliminary evaluation of ideas or hypotheses prior to rigorous testing, not on how a novel hypothesis or idea is first thought up. For a long time, the predominant view among philosophers of discovery was that the initial step of discovery is a mysterious intuitive leap of the human mind that cannot be analyzed further. More recent accounts of discovery informed by evolutionary biology also do not explicate how new ideas are formed. The generation of new ideas is akin to random, blind variations of thought processes, which have to be inspected by the critical mind and assessed as neutral, productive, or useless (Campbell 1960; see also Hull 1988), but the key processes by which new ideas are generated are left unanalyzed.

With the recent rapprochement among philosophy of mind, cognitive science and psychology and the increased integration of empirical research into philosophy of science, these processes have been submitted to closer analysis, and philosophical studies of creativity have seen a surge of interest (e.g. Paul & Kaufman 2014a). The distinctive feature of these studies is that they integrate philosophical analyses with empirical work from cognitive science, psychology, evolutionary biology, and computational neuroscience (Thagard 2012). Analysts have distinguished different kinds and different features of creative thinking and have examined certain features in depth, and from new angles. Recent philosophical research on creativity comprises conceptual analyses and integrated approaches based on the assumption that creativity can be analyzed and that empirical research can contribute to the analysis (Paul & Kaufman 2014b). Two key elements of the cognitive processes involved in creative thinking that have been in the focus of philosophical analysis are analogies ( section 9.2 ) and mental models ( section 9.3 ).

General definitions of creativity highlight novelty or originality and significance or value as distinctive features of a creative act or product (Sternberg & Lubart 1999, Kieran 2014, Paul & Kaufman 2014b, although see Hills & Bird 2019). Different kinds of creativity can be distinguished depending on whether the act or product is novel for a particular individual or entirely novel. Psychologist Margaret Boden distinguishes between psychological creativity (P-creativity) and historical creativity (H-creativity). P-creativity is a development that is new, surprising and important to the particular person who comes up with it. H-creativity, by contrast, is radically novel, surprising, and important—it is generated for the first time (Boden 2004). Further distinctions have been proposed, such as anthropological creativity (construed as a human condition) and metaphysical creativity, a radically new thought or action in the sense that it is unaccounted for by antecedents and available knowledge, and thus constitutes a radical break with the past (Kronfeldner 2009, drawing on Hausman 1984).

Psychological studies analyze the personality traits and creative individuals’ behavioral dispositions that are conducive to creative thinking. They suggest that creative scientists share certain distinct personality traits, including confidence, openness, dominance, independence, introversion, as well as arrogance and hostility. (For overviews of recent studies on personality traits of creative scientists, see Feist 1999, 2006: chapter 5).

Recent work on creativity in philosophy of mind and cognitive science offers substantive analyses of the cognitive and neural mechanisms involved in creative thinking (Abrams 2018, Minai et al 2022) and critical scrutiny of the romantic idea of genius creativity as something deeply mysterious (Blackburn 2014). Some of this research aims to characterize features that are common to all creative processes, such as Thagard and Stewart’s account according to which creativity results from combinations of representations (Thagard & Stewart 2011, but see Pasquale and Poirier 2016). Other research aims to identify the features that are distinctive of scientific creativity as opposed to other forms of creativity, such as artistic creativity or creative technological invention (Simonton 2014).

Many philosophers of science highlight the role of analogy in the development of new knowledge, whereby analogy is understood as a process of bringing ideas that are well understood in one domain to bear on a new domain (Thagard 1984; Holyoak and Thagard 1996). An important source for philosophical thought about analogy is Mary Hesse’s conception of models and analogies in theory construction and development. In this approach, analogies are similarities between different domains. Hesse introduces the distinction between positive, negative, and neutral analogies (Hesse 1966: 8). If we consider the relation between gas molecules and a model for gas, namely a collection of billiard balls in random motion, we will find properties that are common to both domains (positive analogy) as well as properties that can only be ascribed to the model but not to the target domain (negative analogy). There is a positive analogy between gas molecules and a collection of billiard balls because both the balls and the molecules move randomly. There is a negative analogy between the domains because billiard balls are colored, hard, and shiny but gas molecules do not have these properties. The most interesting properties are those properties of the model about which we do not know whether they are positive or negative analogies. This set of properties is the neutral analogy. These properties are the significant properties because they might lead to new insights about the less familiar domain. From our knowledge about the familiar billiard balls, we may be able to derive new predictions about the behavior of gas molecules, which we could then test.

Hesse offers a more detailed analysis of the structure of analogical reasoning through the distinction between horizontal and vertical analogies between domains. Horizontal analogies between two domains concern the sameness or similarity between properties of both domains. If we consider sound and light waves, there are similarities between them: sound echoes, light reflects; sound is loud, light is bright, both sound and light are detectable by our senses. There are also relations among the properties within one domain, such as the causal relation between sound and the loud tone we hear and, analogously, between physical light and the bright light we see. These analogies are vertical analogies. For Hesse, vertical analogies hold the key for the construction of new theories.

Analogies play several roles in science. Not only do they contribute to discovery but they also play a role in the development and evaluation of scientific theories. Current discussions about analogy and discovery have expanded and refined Hesse’s approach in various ways. Some philosophers have developed criteria for evaluating analogy arguments (Bartha 2010). Other work has identified highly significant analogies that were particularly fruitful for the advancement of science (Holyoak and Thagard 1996: 186–188; Thagard 1999: chapter 9). The majority of analysts explore the features of the cognitive mechanisms through which aspects of a familiar domain or source are applied to an unknown target domain in order to understand what is unknown. According to the influential multi-constraint theory of analogical reasoning developed by Holyoak and Thagard, the transfer processes involved in analogical reasoning (scientific and otherwise) are guided or constrained in three main ways: 1) by the direct similarity between the elements involved; 2) by the structural parallels between source and target domain; as well as 3) by the purposes of the investigators, i.e., the reasons why the analogy is considered. Discovery, the formulation of a new hypothesis, is one such purpose.

“In vivo” investigations of scientists reasoning in their laboratories have not only shown that analogical reasoning is a key component of scientific practice, but also that the distance between source and target depends on the purpose for which analogies are sought. Scientists trying to fix experimental problems draw analogies between targets and sources from highly similar domains. In contrast, scientists attempting to formulate new models or concepts draw analogies between less similar domains. Analogies between radically different domains, however, are rare (Dunbar 1997, 2001).

In current cognitive science, human cognition is often explored in terms of model-based reasoning. The starting point of this approach is the notion that much of human reasoning, including probabilistic and causal reasoning as well as problem solving takes place through mental modeling rather than through the application of logic or methodological criteria to a set of propositions (Johnson-Laird 1983; Magnani et al. 1999; Magnani and Nersessian 2002). In model-based reasoning, the mind constructs a structural representation of a real-world or imaginary situation and manipulates this structure. In this perspective, conceptual structures are viewed as models and conceptual innovation as constructing new models through various modeling operations. Analogical reasoning—analogical modeling—is regarded as one of three main forms of model-based reasoning that appear to be relevant for conceptual innovation in science. Besides analogical modeling, visual modeling and simulative modeling or thought experiments also play key roles (Nersessian 1992, 1999, 2009). These modeling practices are constructive in that they aid the development of novel mental models. The key elements of model-based reasoning are the call on knowledge of generative principles and constraints for physical models in a source domain and the use of various forms of abstraction. Conceptual innovation results from the creation of new concepts through processes that abstract and integrate source and target domains into new models (Nersessian 2009).

Some critics have argued that despite the large amount of work on the topic, the notion of mental model is not sufficiently clear. Thagard seeks to clarify the concept by characterizing mental models in terms of neural processes (Thagard 2010). In his approach, mental models are produced through complex patterns of neural firing, whereby the neurons and the interconnections between them are dynamic and changing. A pattern of firing neurons is a representation when there is a stable causal correlation between the pattern or activation and the thing that is represented. In this research, questions about the nature of model-based reasoning are transformed into questions about the brain mechanisms that produce mental representations.

The above sections again show that the study of scientific discovery integrates different approaches, combining conceptual analysis of processes of knowledge generation with empirical work on creativity, drawing heavily and explicitly on current research in psychology and cognitive science, and on in vivo laboratory observations, as well as brain imaging techniques (Kounios & Beeman 2009, Thagard & Stewart 2011).

Earlier critics of AI-based theories of scientific discoveries argued that a computer cannot devise new concepts but is confined to the concepts included in the given computer language (Hempel 1985: 119–120). It cannot design new experiments, instruments, or methods. Subsequent computational research on scientific discovery was driven by the motivation to contribute computational tools to aid scientists in their research (Addis et al. 2016). It appears that computational methods can be used to generate new results leading to refereed scientific publications in astrophysics, cancer research, ecology, and other fields (Langley 2000). However, the philosophical discussion has continued about the question of whether these methods really generate new knowledge or whether they merely speed up data processing. It is also still an open question whether data-intensive science is fundamentally different from traditional research, for instance regarding the status of hypothesis or theory in data-intensive research (Pietsch 2015).

In the wake of recent developments in machine learning, some older discussions about automated discovery have been revived. The availability of vastly improved computational tools and software for data analysis has stimulated new discussions about computer-generated discovery (see Leonelli 2020). It is largely uncontroversial that machine learning tools can aid discovery, for instance in research on antibiotics (Stokes et al, 2020). The notion of “robot scientist” is mostly used metaphorically, and the vision that human scientists may one day be replaced by computers – by successors of the laboratory automation systems “Adam” and “Eve”, allegedly the first “robot scientists” – is evoked in writings for broader audiences (see King et al. 2009, Williams et al. 2015, for popularized descriptions of these systems), although some interesting ethical challenges do arise from “superhuman AI” (see Russell 2021). It also appears that, on the notion that products of creative acts are both novel and valuable, AI systems should be called “creative,” an implication which not all analysts will find plausible (Boden 2014)

Philosophical analyses focus on various questions arising from the processes involving human-machine complexes. One issue relevant to the problem of scientific discovery arises from the opacity of machine learning. If machine learning indeed escapes human understanding, how can we be warranted to say that knowledge or understanding is generated by deep learning tools? Might we have reason to say that humans and machines are “co-developers” of knowledge (Tamaddoni-Nezhad et al. 2021)?

New perspectives on scientific discovery have also opened up in the context of social epistemology (see Goldman & O’Connor 2021). Social epistemology investigates knowledge production as a group process, specifically the epistemic effects of group composition in terms of cognitive diversity and unity and social interactions within groups or institutions such as testimony and trust, peer disagreement and critique, and group justification, among others. On this view, discovery is a collective achievement, and the task is to explore how assorted social-epistemic activities or practices have an impact on the knowledge generated by groups in question. There are obvious implications for debates about scientific discovery of recent research in the different branches of social epistemology. Social epistemologists have examined individual cognitive agents in their roles as group members (as providers of information or as critics) and the interactions among these members (Longino 2001), groups as aggregates of diverse agents, or the entire group as epistemic agent (e.g., Koons 2021, Dragos 2019).

Standpoint theory, for instance, explores the role of outsiders in knowledge generation, considering how the sociocultural structures and practices in which individuals are embedded aid or obstruct the generation of creative ideas. According to standpoint theorists, people with standpoint are politically aware and politically engaged people outside the mainstream. Because people with standpoint have different experiences and access to different domains of expertise than most members of a culture, they can draw on rich conceptual resources for creative thinking (Solomon 2007).

Social epistemologists examining groups as aggregates of agents consider to what extent diversity among group members is conducive to knowledge production and whether and to what extent beliefs and attitudes must be shared among group members to make collective knowledge possible (Bird 2014). This is still an open question. Some formal approaches to model the influence of diversity on knowledge generation suggest that cognitive diversity is beneficial to collective knowledge generation (Weisberg and Muldoon 2009), but others have criticized the model (Alexander et al (2015), see also Thoma (2015) and Poyhönen (2017) for further discussion).

This essay has illustrated that philosophy of discovery has come full circle. Philosophy of discovery has once again become a thriving field of philosophical study, now intersecting with, and drawing on philosophical and empirical studies of creative thinking, problem solving under uncertainty, collective knowledge production, and machine learning. Recent approaches to discovery are typically explicitly interdisciplinary and integrative, cutting across previous distinctions among hypothesis generation and theory building, data collection, assessment, and selection; as well as descriptive-analytic, historical, and normative perspectives (Danks & Ippoliti 2018, Michel 2021). The goal no longer is to provide one overarching account of scientific discovery but to produce multifaceted analyses of past and present activities of knowledge generation in all their complexity and heterogeneity that are illuminating to the non-scientist and the scientific researcher alike.

• Abraham, A. 2019, The Neuroscience of Creativity, Cambridge: Cambridge University Press.
• Addis, M., Sozou, P.D., Gobet, F. and Lane, P. R., 2016, “Computational scientific discovery and cognitive science theories”, in Mueller, V. C. (ed.) Computing and Philosophy , Springer, 83–87.
• Alexander, J., Himmelreich, J., and Thompson, C. 2015, Epistemic Landscapes, Optimal Search, and the Division of Cognitive Labor, Philosophy of Science 82: 424–453.
• Arabatzis, T. 1996, “Rethinking the ‘Discovery’ of the Electron,” Studies in History and Philosophy of Science Part B Studies In History and Philosophy of Modern Physics , 27: 405–435.
• Bartha, P., 2010, By Parallel Reasoning: The Construction and Evaluation of Analogical Arguments , New York: Oxford University Press.
• Bechtel, W. and R. Richardson, 1993, Discovering Complexity , Princeton: Princeton University Press.
• Benjamin, A.C., 1934, “The Mystery of Scientific Discovery ” Philosophy of Science , 1: 224–36.
• Bird, A. 2014, “When is There a Group that Knows? Distributed Cognition, Scientific Knowledge, and the Social Epistemic Subject”, in J. Lackey (ed.), Essays in Collective Epistemology , Oxford: Oxford University Press, 42–63.
• Blackburn, S. 2014, “Creativity and Not-So-Dumb Luck”, in Paul, E. S. and Kaufman, S. B. (eds.), The Philosophy of Creativity: New Essays , New York: Oxford Academic online edn. https://doi.org/10.1093/acprof:oso/9780199836963.003.0008.
• Blackwell, R.J., 1969, Discovery in the Physical Sciences , Notre Dame: University of Notre Dame Press.
• Boden, M.A., 2004, The Creative Mind: Myths and Mechanisms , London: Routledge.
• –––, 2014, “Creativity and Artificial Intelligence: A Contradiction in Terms?”, in Paul, E. S. and Kaufman, S. B. (eds.), The Philosophy of Creativity: New Essays (New York: Oxford Academic online edn., https://doi.org/10.1093/acprof:oso/9780199836963.003.0012 .
• Brannigan, A., 1981, The Social Basis of Scientific Discoveries , Cambridge: Cambridge University Press.
• Brem, S. and L.J. Rips, 2000, “Explanation and Evidence in Informal Argument”, Cognitive Science , 24: 573–604.
• Campbell, D., 1960, “Blind Variation and Selective Retention in Creative Thought as in Other Knowledge Processes”, Psychological Review , 67: 380–400.
• Carmichael, R.D., 1922, “The Logic of Discovery”, The Monist , 32: 569–608.
• –––, 1930, The Logic of Discovery , Chicago: Open Court.
• Chow, S. 2015, “Many Meanings of ‘Heuristic’”, British Journal for the Philosophy of Science , 66: 977–1016
• Craver, C.F., 2002, “Interlevel Experiments, Multilevel Mechanisms in the Neuroscience of Memory”, Philosophy of Science Supplement , 69: 83–97.
• Craver, C.F. and L. Darden, 2013, In Search of Mechanisms: Discoveries across the Life Sciences , Chicago: University of Chicago Press.
• Curd, M., 1980, “The Logic of Discovery: An Analysis of Three Approaches”, in T. Nickles (ed.) Scientific Discovery, Logic, and Rationality , Dordrecht: D. Reidel, 201–19.
• Danks, D. & Ippoliti, E. (eds.) 2018, Building Theories: Heuristics and Hypotheses in Sciences , Cham: Springer.
• Darden, L., 1991, Theory Change in Science: Strategies from Mendelian Genetics , New York: Oxford University Press.
• –––, 2002, “Strategies for Discovering Mechanisms: Schema Instantiation, Modular Subassembly, Forward/Backward Chaining”, Philosophy of Science , 69: S354-S65.
• –––, 2009, “Discovering Mechanisms in Molecular Biology: Finding and Fixing Incompleteness and Incorrectness”, in J. Meheus and T. Nickles (eds.), Models of Discovery and Creativity , Dordrecht: Springer, 43–55.
• Dewey, J. 1910, How We Think . Boston: D.C. Heath
• Dragos, C., 2019, “Groups Can Know How” American Philosophical Quarterly 56: 265–276
• Ducasse, C.J., 1951, “Whewell’s Philosophy of Scientific Discovery II”, The Philosophical Review , 60(2): 213–34.
• Dunbar, K., 1997, “How scientists think: On-line creativity and conceptual change in science”, in T.B. Ward, S.M. Smith, and J. Vaid (eds.), Conceptual Structures and Processes: Emergence, Discovery, and Change , Washington, DC: American Psychological Association Press, 461–493.
• –––, 2001, “The Analogical Paradox: Why Analogy is so Easy in Naturalistic Settings Yet so Difficult in Psychological Laboratories”, in D. Gentner, K.J. Holyoak, and B.N. Kokinov (eds.), The Analogical Mind: Perspectives from Cognitive Science , Cambridge, MA: MIT Press.
• Dunbar, K, J. Fugelsang, and C Stein, 2007, “Do Naïve Theories Ever Go Away? Using Brain and Behavior to Understand Changes in Concepts”, in M. Lovett and P. Shah (eds.), Thinking with Data: 33rd Carnegie Symposium on Cognition , Mahwah: Erlbaum, 193–205.
• Feist, G.J., 1999, “The Influence of Personality on Artistic and Scientific Creativity”, in R.J. Sternberg (ed.), Handbook of Creativity , New York: Cambridge University Press, 273–96.
• –––, 2006, The psychology of science and the origins of the scientific mind , New Haven: Yale University Press.
• Gillies D., 1996, Artificial intelligence and scientific method . Oxford: Oxford University Press.
• –––, 2018 “Discovering Cures in Medicine” in Danks, D. & Ippoliti, E. (eds.), Building Theories: Heuristics and Hypotheses in Sciences , Cham: Springer, 83–100.
• Goldman, Alvin & O’Connor, C., 2021, “Social Epistemology”, The Stanford Encyclopedia of Philosophy (Winter 2021 Edition), Edward N. Zalta (ed.), URL = <https://plato.stanford.edu/archives/win2021/entries/epistemology-social/>.
• Gramelsberger, G. 2011, “What Do Numerical (Climate) Models Really Represent?” Studies in History and Philosophy of Science 42: 296–302.
• Gutting, G., 1980, “Science as Discovery”, Revue internationale de philosophie , 131: 26–48.
• Hanson, N.R., 1958, Patterns of Discovery , Cambridge: Cambridge University Press.
• –––, 1960, “Is there a Logic of Scientific Discovery?”, Australasian Journal of Philosophy , 38: 91–106.
• –––, 1965, “Notes Toward a Logic of Discovery”, in R.J. Bernstein (ed.), Perspectives on Peirce. Critical Essays on Charles Sanders Peirce , New Haven and London: Yale University Press, 42–65.
• Harman, G.H., 1965, “The Inference to the Best Explanation”, Philosophical Review , 74.
• Hausman, C. R. 1984, A Discourse on Novelty and Creation , New York: SUNY Press.
• Hempel, C.G., 1985, “Thoughts in the Limitations of Discovery by Computer”, in K. Schaffner (ed.), Logic of Discovery and Diagnosis in Medicine , Berkeley: University of California Press, 115–22.
• Hesse, M., 1966, Models and Analogies in Science , Notre Dame: University of Notre Dame Press.
• Hey, S. 2016 “Heuristics and Meta-heuristics in Scientific Judgement”, British Journal for the Philosophy of Science , 67: 471–495
• Hills, A., Bird, A. 2019, “Against Creativity”, Philosophy and Phenomenological Research , 99: 694–713.
• Holyoak, K.J. and P. Thagard, 1996, Mental Leaps: Analogy in Creative Thought , Cambridge, MA: MIT Press.
• Howard, D., 2006, “Lost Wanderers in the Forest of Knowledge: Some Thoughts on the Discovery-Justification Distinction”, in J. Schickore and F. Steinle (eds.), Revisiting Discovery and Justification. Historical and Philosophical Perspectives on the Context Distinction , Dordrecht: Springer, 3–22.
• Hoyningen-Huene, P., 1987, “Context of Discovery and Context of Justification”, Studies in History and Philosophy of Science , 18: 501–15.
• Hull, D.L., 1988, Science as Practice: An Evolutionary Account of the Social and Conceptual Development of Science , Chicago: University of Chicago Press.
• Ippoliti, E. 2018, “Heuristic Logic. A Kernel” in Danks, D. & Ippoliti, E. (eds.) Building Theories: Heuristics and Hypotheses in Sciences , Cham: Springer, 191–212
• Jantzen, B.C., 2016, “Discovery without a ‘Logic’ would be a Miracle”, Synthese , 193: 3209–3238.
• Johnson-Laird, P., 1983, Mental Models , Cambridge: Cambridge University Press.
• Kieran, M., 2014, “Creativity as a Virtue of Character,” in E. Paul and S. B. Kaufman (eds.), The Philosophy of Creativity: New Essays . Oxford: Oxford University Press, 125–44
• King, R. D. et al. 2009, “The Automation of Science”, Science 324: 85–89.
• Koehler, D.J., 1991, “Explanation, Imagination, and Confidence in Judgment”, Psychological Bulletin , 110: 499–519.
• Koertge, N. 1980, “Analysis as a Method of Discovery during the Scientific Revolution” in Nickles, T. (ed.) Scientific Discovery, Logic, and Rationality vol. I, Dordrecht: Reidel, 139–157
• Koons, J.R. 2021, “Knowledge as a Collective Status”, Analytic Philosophy , https://doi.org/10.1111/phib.12224
• Kounios, J. and Beeman, M. 2009, “The Aha! Moment : The Cognitive Neuroscience of Insight”, Current Directions in Psychological Science , 18: 210–16.
• Kordig, C., 1978, “Discovery and Justification”, Philosophy of Science , 45: 110–17.
• Kronfeldner, M. 2009, “Creativity Naturalized”, The Philosophical Quarterly 59: 577–592.
• Kuhn, T.S., 1970 [1962], The Structure of Scientific Revolutions , 2 nd edition, Chicago: The University of Chicago Press; first edition, 1962.
• Kulkarni, D. and H.A. Simon, 1988, “The processes of scientific discovery: The strategy of experimentation”, Cognitive Science , 12: 139–76.
• Langley, P., 2000, “The Computational Support of Scientific Discovery”, International Journal of Human-Computer Studies , 53: 393–410.
• Langley, P., H.A. Simon, G.L. Bradshaw, and J.M. Zytkow, 1987, Scientific Discovery: Computational Explorations of the Creative Processes , Cambridge, MA: MIT Press.
• Laudan, L., 1980, “Why Was the Logic of Discovery Abandoned?” in T. Nickles (ed.), Scientific Discovery (Volume I), Dordrecht: D. Reidel, 173–83.
• Leonelli, S. 2020, “Scientific Research and Big Data”, The Stanford Encyclopedia of Philosophy (Summer 2020 Edition), Edward N. Zalta (ed.), URL = <https://plato.stanford.edu/archives/sum2020/entries/science-big-data/>
• Leplin, J., 1987, “The Bearing of Discovery on Justification”, Canadian Journal of Philosophy , 17: 805–14.
• Longino, H. 2001, The Fate of Knowledge , Princeton: Princeton University Press
• Lugg, A., 1985, “The Process of Discovery”, Philosophy of Science , 52: 207–20.
• Magnani, L., 2000, Abduction, Reason, and Science: Processes of Discovery and Explanation , Dordrecht: Kluwer.
• –––, 2009, “Creative Abduction and Hypothesis Withdrawal”, in J. Meheus and T. Nickles (eds.), Models of Discovery and Creativity , Dordrecht: Springer.
• Magnani, L. and N.J. Nersessian, 2002, Model-Based Reasoning: Science, Technology, and Values , Dordrecht: Kluwer.
• Magnani, L., N.J. Nersessian, and P. Thagard, 1999, Model-Based Reasoning in Scientific Discovery , Dordrecht: Kluwer.
• Michel, J. (ed.) 2021, Making Scientific Discoveries. Interdisciplinary Reflections , Brill | mentis.
• Minai, A., Doboli, S., Iyer, L. 2022 “Models of Creativity and Ideation: An Overview” in Ali A. Minai, Jared B. Kenworthy, Paul B. Paulus, Simona Doboli (eds.), Creativity and Innovation. Cognitive, Social, and Computational Approaches , Springer, 21–46.
• Nersessian, N.J., 1992, “How do scientists think? Capturing the dynamics of conceptual change in science”, in R. Giere (ed.), Cognitive Models of Science , Minneapolis: University of Minnesota Press, 3–45.
• –––, 1999, “Model-based reasoning in conceptual change”, in L. Magnani, N.J. Nersessian and P. Thagard (eds.), Model-Based Reasoning in Scientific Discovery , New York: Kluwer, 5–22.
• –––, 2009, “Conceptual Change: Creativity, Cognition, and Culture ” in J. Meheus and T. Nickles (eds.), Models of Discovery and Creativity , Dordrecht: Springer, 127–66.
• Newell, A. and H. A Simon, 1971, “Human Problem Solving: The State of the Theory in 1970”, American Psychologist , 26: 145–59.
• Newton, I. 1718, Opticks; or, A Treatise of the Reflections, Inflections and Colours of Light , London: Printed for W. and J. Innys, Printers to the Royal Society.
• Nickles, T., 1984, “Positive Science and Discoverability”, PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association , 1984: 13–27.
• –––, 1985, “Beyond Divorce: Current Status of the Discovery Debate”, Philosophy of Science , 52: 177–206.
• –––, 1989, “Truth or Consequences? Generative versus Consequential Justification in Science”, PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association , 1988, 393–405.
• –––, 2018, “TTT: A Fast Heuristic to New Theories?” in Danks, D. & Ippoliti, E. (eds.) Building Theories: Heuristics and Hypotheses in Sciences , Cham: Springer, 213–244.
• Pasquale, J.-F. de and Poirier, P. 2016, “Convolution and Modal Representations in Thagard and Stewart’s Neural Theory of Creativity: A Critical Analysis ”, Synthese , 193: 1535–1560
• Paul, E. S. and Kaufman, S. B. (eds.), 2014a, The Philosophy of Creativity: New Essays , New York: Oxford Academic online edn., https://doi.org/10.1093/acprof:oso/9780199836963.001.0001.
• –––, 2014b, “Introducing: The Philosophy of Creativity”, in Paul, E. S. and Kaufman, S. B. (eds.), The Philosophy of Creativity: New Essays (New York: Oxford Academic online edn., https://doi.org/10.1093/acprof:oso/9780199836963.003.0001.
• Pietsch, W. 2015, “Aspects of Theory-Ladenness in Data-Intensive Science”, Philosophy of Science 82: 905–916.
• Popper, K., 2002 [1934/1959], The Logic of Scientific Discovery , London and New York: Routledge; original published in German in 1934; first English translation in 1959.
• Pöyhönen, S. 2017, “Value of Cognitive Diversity in Science”, Synthese , 194(11): 4519–4540. doi:10.1007/s11229–016-1147-4
• Pulte, H. 2019, “‘‘Tis Much Better to Do a Little with Certainty’: On the Reception of Newton’s Methodology”, in The Reception of Isaac Newton in Europe , Pulte, H, and Mandelbrote, S. (eds.), Continuum Publishing Corporation, 355–84.
• Reichenbach, H., 1938, Experience and Prediction. An Analysis of the Foundations and the Structure of Knowledge , Chicago: The University of Chicago Press.
• Richardson, A., 2006, “Freedom in a Scientific Society: Reading the Context of Reichenbach’s Contexts”, in J. Schickore and F. Steinle (eds.), Revisiting Discovery and Justification. Historical and Philosophical Perspectives on the Context Distinction , Dordrecht: Springer, 41–54.
• Russell, S. 2021, “Human-Compatible Artificial Intelligence”, in Human Like Machine Intelligence , Muggleton, S. and Charter, N. (eds.), Oxford: Oxford University Press, 4–23
• Schaffer, S., 1986, “Scientific Discoveries and the End of Natural Philosophy”, Social Studies of Science , 16: 387–420.
• –––, 1994, “Making Up Discovery”, in M.A. Boden (ed.), Dimensions of Creativity , Cambridge, MA: MIT Press, 13–51.
• Schaffner, K., 1993, Discovery and Explanation in Biology and Medicine , Chicago: University of Chicago Press.
• –––, 2008 “Theories, Models, and Equations in Biology: The Heuristic Search for Emergent Simplifications in Neurobiology”, Philosophy of Science , 75: 1008–21.
• Schickore, J. and F. Steinle, 2006, Revisiting Discovery and Justification. Historical and Philosophical Perspectives on the Context Distinction , Dordrecht: Springer.
• Schiller, F.C.S., 1917, “Scientific Discovery and Logical Proof”, in C.J. Singer (ed.), Studies in the History and Method of Science (Volume 1), Oxford: Clarendon, 235–89.
• Simon, H.A., 1973, “Does Scientific Discovery Have a Logic?”, Philosophy of Science , 40: 471–80.
• –––, 1977, Models of Discovery and Other Topics in the Methods of Science , Dordrecht: D. Reidel.
• Simon, H.A., P.W. Langley, and G.L. Bradshaw, 1981, “Scientific Discovery as Problem Solving”, Synthese , 47: 1–28.
• Smith, G.E., 2002, “The Methodology of the Principia ”, in G.E. Smith and I.B. Cohen (eds), The Cambridge Companion to Newton , Cambridge: Cambridge University Press, 138–73.
• Simonton, D. K., “Hierarchies of Creative Domains: Disciplinary Constraints on Blind Variation and Selective Retention”, in Paul, E. S. and Kaufman, S. B. (eds), The Philosophy of Creativity: New Essays , New York: Oxford Academic online edn. https://doi.org/10.1093/acprof:oso/9780199836963.003.0013
• Snyder, L.J., 1997, “Discoverers’ Induction”, Philosophy of Science , 64: 580–604.
• Solomon, M., 2009, “Standpoint and Creativity”, Hypatia : 226–37.
• Sternberg, R J. and T. I. Lubart, 1999, “The concept of creativity: Prospects and paradigms,” in R. J. Sternberg (ed.) Handbook of Creativity , Cambridge: Cambridge University Press, 3–15.
• Stokes, D., 2011, “Minimally Creative Thought”, Metaphilosophy , 42: 658–81.
• Tamaddoni-Nezhad, A., Bohan, D., Afroozi Milani, G., Raybould, A., Muggleton, S., 2021, “Human–Machine Scientific Discovery”, in Human Like Machine Intelligence , Muggleton, S. and Charter, N., (eds.), Oxford: Oxford University Press, 297–315
• Thagard, P., 1984, “Conceptual Combination and Scientific Discovery”, PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association , 1984(1): 3–12.
• –––, 1999, How Scientists Explain Disease , Princeton: Princeton University Press.
• –––, 2010, “How Brains Make Mental Models”, in L. Magnani, N.J. Nersessian and P. Thagard (eds.), Model-Based Reasoning in Science & Technology , Berlin and Heidelberg: Springer, 447–61.
• –––, 2012, The Cognitive Science of Science , Cambridge, MA: MIT Press.
• Thagard, P. and Stewart, T. C., 2011, “The AHA! Experience: Creativity Through Emergent Binding in Neural Networks”, Cognitive Science , 35: 1–33.
• Thoma, Johanna, 2015, “The Epistemic Division of Labor Revisited”, Philosophy of Science , 82: 454–472. doi:10.1086/681768
• Weber, M., 2005, Philosophy of Experimental Biology , Cambridge: Cambridge University Press.
• Whewell, W., 1996 [1840], The Philosophy of the Inductive Sciences (Volume II), London: Routledge/Thoemmes.
• Weisberg, M. and Muldoon, R., 2009, “Epistemic Landscapes and the Division of Cognitive Labor”, Philosophy of Science , 76: 225–252. doi:10.1086/644786
• Williams, K. et al. 2015, “Cheaper Faster Drug Development Validated by the Repositioning of Drugs against Neglected Tropical Diseases”, Journal of the Royal Society Interface 12: 20141289. http://dx.doi.org/10.1098/rsif.2014.1289.
• Zahar, E., 1983, “Logic of Discovery or Psychology of Invention?”, British Journal for the Philosophy of Science , 34: 243–61.
• Zednik, C. and Jäkel, F. 2016 “Bayesian Reverse-Engineering Considered as a Research Strategy for Cognitive Science”, Synthese , 193, 3951–3985.

How to cite this entry . Preview the PDF version of this entry at the Friends of the SEP Society . Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO). Enhanced bibliography for this entry at PhilPapers , with links to its database.

[Please contact the author with suggestions.]

abduction | analogy and analogical reasoning | cognitive science | epistemology: social | knowledge: analysis of | Kuhn, Thomas | models in science | Newton, Isaac: Philosophiae Naturalis Principia Mathematica | Popper, Karl | rationality: historicist theories of | scientific method | scientific research and big data | Whewell, William

Copyright © 2022 by Jutta Schickore < jschicko @ indiana . edu >

• Accessibility

Support SEP

Mirror sites.

View this site from another server:

• Info about mirror sites

The Stanford Encyclopedia of Philosophy is copyright © 2023 by The Metaphysics Research Lab , Department of Philosophy, Stanford University

Library of Congress Catalog Data: ISSN 1095-5054

Environment

Understanding our justification systems, five key concepts that stem from the justification hypothesis.

Posted November 1, 2013

“I am too stupid for that,” the woman (I’ll call her Annie) said to me after I asked her what prevented her from going on to pursue a college degree. I asked this question in the context of a psychiatric evaluation I was doing. Annie had been hospitalized following an overdose that she genuinely thought would kill her. A standardized assessment of her intellect had documented that her IQ was in the High Average Range, and in no way shape or form was she “stupid.” But that is how she saw herself. In fact, she saw herself as weak or lacking in most every respect. Although she was of average attractiveness , she saw herself as ugly. Although she could play musical instruments, she saw herself as having no talents. Indeed, she was almost relentless in seeing herself as worthless and inferior. Diagnostically, she met criteria for an Avoidant Personality Disorder and a “double depression .”

I was doing this evaluation at a time when I was thinking a lot about human self-consciousness. As part of my dissertation, I was studying cognitive biases and the fact that most people self-enhance and seek to paint themselves in the most rosy picture the data can justify. But although true of most, people with depression tended not to do this, or at least do it less, with some being grossly self-critical. I was also getting my first real deep exposure to modern psychodynamic theory, and learning about how the self-consciousness system often filters out subconscious motivation . In the language of dynamic defense mechanisms , the woman was “turning against the self.” which is often thought to be a way of redirecting aggression that is dangerous to express. Ultimately, my study of evolutionary theory, social and cognitive psychology, and psychodynamic theory intersected with hearing Annie’s story, and a light bulb went off that allowed me to realize that self-enhancement and turning against the self might be both part of the same basic processes. The light bulb would reveal a key design feature of human self-consciousness.

One of the most obvious features of Annie’s story was her family history. In what is an all-to-common story line to those who work in psychiatric settings, this woman’s father was a domineering and abusive man, who ruled the household with an iron fist. Her mother was a timid woman, who obviously walked on eggshells. Interestingly, although her father was verbally abusive toward Annie, he was never physically abusive. But he was physically abusive to her older brother. Indeed, she had several distinctive memories of her father beating her brother, telling her brother that he ought to be more like his obedient sister. Upon hearing that aspect of her story, something clicked and, over time, a whole new understanding of self-consciousness emerged in me.

At a primal level, this woman watched her brother be beaten, and yet it was a fate she always somehow avoided. How did she avoid it? One very plausible answer is by turning against herself. As her father behaved the way he did, a natural response would to become angry and defiant and challenging. And, yet, doing so would be highly dangerous. If, on the other hand, she saw herself as weak, unworthy, and undeserving, these beliefs justify submission and deference, which influenced her father’s behavior, namely by avoiding an attack.

In this light then, Annie’s sense that she was lesser than others perhaps stemmed from the same basic process that “normals” engage in when they moderately self-enhance. By thinking that we are a bit smarter, more attractive, or more effective than others, surely that places us in a position to frame what we do in a more influential way. In short, although very different in content, both self-criticalness and self-enhancement might tie into the process of justifying one’s actions to one’s self in a social context. Ultimately, the exchange with Annie led me to formulate the “ Justification Hypothesis ” which is the idea that our self-consciousness system emerged in the evolutionary landscape as a social reason giving system. That is, we humans were not shaped by evolution for general abstract reasoning, but our reasoning capacities emerged in the context of a social environment where the adaptive task was to generate social reasons for our actions.

Although there are many different directions the Justification Hypothesis can take us, there are five general key concepts that fall from this analysis that I believe people everywhere can benefit from understanding.

1. Language-based beliefs and values are organized into systems of justification . Beliefs and values are not randomly distributed in people’s heads, but instead are networked together to form systems that ultimately function to legitimize action.

2. Justification systems exist at the individual and societal level, and the two are connected . I have my justifications for why I am the way I am. I can explain to my kids why they should follow what I say or to my colleagues why my ideas are valid. At the same time, large scale justifications provide the glue that gives a society a shared identity and worldview in a particular context. I reason the way I do in a way that is deeply connected to my sociocultural context. Think, for example, of how justifications regarding gender roles are different in Canada as opposed to Saudi Arabi, or how they compare today relative to 100 years ago. We exist in a sea of justification and carve out our individual path in that context.

3. Humans have two streams of consciousness, experiencing and justifying. The justifying capacity of humans is a late evolutionary add on. Prior to it, others animals have been navigating the environment for eons via sensory-perceptual-emotional processes of action investment. Thus, these are two very different systems, the mutual flow of which makes up human consciousness in its totality.

4. For adults, there are two domains of justification, the private and the public. As children, we download the reason giving systems of the social context in which we are born. As our capacities develop, we begin to engage in reasoning on our own. Privately, we must legitimize our own thoughts and develop narrative for who we are and why we do what we do. And, our family, friends, peers, etc. will often want to be informed about what we think and why we do what we do.

5. There is filtering between the experiential, private and public justification systems. In the film Liar Liar , Jim Carey plays a cheese ball lawyer who always is “BSing” his friends, adversaries and family.

The premise of the movie is that his son gets so frustrated by this he wishes his father could not tell a lie for 24 hours, and the movie is largely about the comical situational analysis of the character as the boy’s wish miraculously comes true and the lawyer must say exactly what his private narrator thinks. As such, the movie is a nice example of the private to public filter. Indeed, we all have experience with this, which can be noticed with the question, “What if everyone of your thoughts became public to everyone?” I don’t know anyone who doesn’t find that to be a somewhat troubling thought. The private to public filter is called the Rogerian filter because we generally filter to avoid injuring others in a way that will evoke problematic reactions and judgments.

But, as Freud so clearly noted, we also filter out feelings, images, wishes (experiential processes) from our self-conscious system.

In his book Ego Defenses and the Legitimization of Behavior, Swanson (1988) made exactly this point, explicitly arguing that we should think of all ego defenses as “justifications that people make to themselves and others—justifications so designed that the defender, not just other people, can accept them.” Think back to Annie. It seems highly likely that she had hostile feelings toward her father. However, if those created anxiety , then the belief that she was deserving in some way of his abuse would block her from those feelings. They would be filtered out, or repressed and her system would be returned to equilibrium by the rationalization that she was worthless.

The following diagram depicts the three domains of consciousness and the filters.

I believe that justification systems are one of the most central aspects of human psychology to be understood. I hope this blog jives with your justification system and you find these insights foster a greater understanding of the nature of your justification system and how it relates to your experiential mind and the socio-cultural context in which you live.

Gregg Henriques, Ph.D. , is a professor of psychology at James Madison University.

• Find a Therapist
• Find a Treatment Center
• Find a Psychiatrist
• Find a Support Group
• Find Online Therapy
• United States
• Brooklyn, NY
• Chicago, IL
• Houston, TX
• Los Angeles, CA
• New York, NY
• Portland, OR
• San Diego, CA
• San Francisco, CA
• Seattle, WA
• Washington, DC
• Asperger's
• Bipolar Disorder
• Chronic Pain
• Eating Disorders
• Passive Aggression
• Personality
• Goal Setting
• Positive Psychology
• Stopping Smoking
• Low Sexual Desire
• Relationships
• Child Development
• Self Tests NEW
• Therapy Center
• Diagnosis Dictionary
• Types of Therapy

When we fall prey to perfectionism, we think we’re honorably aspiring to be our very best, but often we’re really just setting ourselves up for failure, as perfection is impossible and its pursuit inevitably backfires.

• Emotional Intelligence
• Gaslighting
• Affective Forecasting
• Neuroscience

Justifying the Justification Hypothesis: scientific-humanism, Equilintegration (EI) Theory, and the Beliefs, Events, and Values Inventory (BEVI)

Affiliation.

• 1 James Madison University, Harrisonburg, VA 22807, USA. [email protected]
• PMID: 15558624
• DOI: 10.1002/jclp.20092

The Justification Hypothesis (JH; Henriques, 2003) is a basic, general, and macro-level construct that is highly compelling. However, it needs greater specification (i.e., justification) regarding what it is, how it might be operationalized and measured, and what it does and does not predict in the real world. In the present analysis, the act of "justification" is conceptualized as the ongoing attempt to convince self and/or others that one's beliefs and values, which is to say one's "version of reality" or VOR, is correct, defensible, and good. In addressing these issues, this paper is divided into two complementary parts: (a) consideration of justification dynamics and exemplars from a scientific-humanist perspective and (b) an examination of how justification systems and processes have been studied vis-a-vis research and theory on beliefs and values as well as an extant model--Equilintegration (EI) Theory--and method--the Beliefs, Events, and Values Inventory (BEVI).

• Life Change Events*
• Psychological Theory*
• Social Values*
• Surveys and Questionnaires*

• Search Menu
• Sign in through your institution
• Browse content in Arts and Humanities
• Browse content in Archaeology
• Anglo-Saxon and Medieval Archaeology
• Archaeological Methodology and Techniques
• Archaeology by Region
• Archaeology of Religion
• Archaeology of Trade and Exchange
• Biblical Archaeology
• Contemporary and Public Archaeology
• Environmental Archaeology
• Historical Archaeology
• History and Theory of Archaeology
• Industrial Archaeology
• Landscape Archaeology
• Mortuary Archaeology
• Prehistoric Archaeology
• Underwater Archaeology
• Urban Archaeology
• Zooarchaeology
• Browse content in Architecture
• Architectural Structure and Design
• History of Architecture
• Residential and Domestic Buildings
• Theory of Architecture
• Browse content in Art
• Art Subjects and Themes
• History of Art
• Industrial and Commercial Art
• Theory of Art
• Biographical Studies
• Byzantine Studies
• Browse content in Classical Studies
• Classical History
• Classical Philosophy
• Classical Mythology
• Classical Numismatics
• Classical Literature
• Classical Reception
• Classical Art and Architecture
• Classical Oratory and Rhetoric
• Greek and Roman Papyrology
• Greek and Roman Epigraphy
• Greek and Roman Law
• Greek and Roman Archaeology
• Late Antiquity
• Religion in the Ancient World
• Social History
• Digital Humanities
• Browse content in History
• Colonialism and Imperialism
• Diplomatic History
• Environmental History
• Genealogy, Heraldry, Names, and Honours
• Genocide and Ethnic Cleansing
• Historical Geography
• History by Period
• History of Emotions
• History of Agriculture
• History of Education
• History of Gender and Sexuality
• Industrial History
• Intellectual History
• International History
• Labour History
• Legal and Constitutional History
• Local and Family History
• Maritime History
• Military History
• National Liberation and Post-Colonialism
• Oral History
• Political History
• Public History
• Regional and National History
• Revolutions and Rebellions
• Slavery and Abolition of Slavery
• Social and Cultural History
• Theory, Methods, and Historiography
• Urban History
• World History
• Browse content in Language Teaching and Learning
• Language Learning (Specific Skills)
• Language Teaching Theory and Methods
• Browse content in Linguistics
• Applied Linguistics
• Cognitive Linguistics
• Computational Linguistics
• Forensic Linguistics
• Grammar, Syntax and Morphology
• Historical and Diachronic Linguistics
• History of English
• Language Evolution
• Language Reference
• Language Acquisition
• Language Variation
• Language Families
• Lexicography
• Linguistic Anthropology
• Linguistic Theories
• Linguistic Typology
• Phonetics and Phonology
• Psycholinguistics
• Sociolinguistics
• Translation and Interpretation
• Writing Systems
• Browse content in Literature
• Bibliography
• Children's Literature Studies
• Literary Studies (Romanticism)
• Literary Studies (American)
• Literary Studies (Asian)
• Literary Studies (European)
• Literary Studies (Eco-criticism)
• Literary Studies (Modernism)
• Literary Studies - World
• Literary Studies (1500 to 1800)
• Literary Studies (19th Century)
• Literary Studies (20th Century onwards)
• Literary Studies (African American Literature)
• Literary Studies (British and Irish)
• Literary Studies (Early and Medieval)
• Literary Studies (Fiction, Novelists, and Prose Writers)
• Literary Studies (Gender Studies)
• Literary Studies (Graphic Novels)
• Literary Studies (History of the Book)
• Literary Studies (Plays and Playwrights)
• Literary Studies (Poetry and Poets)
• Literary Studies (Postcolonial Literature)
• Literary Studies (Queer Studies)
• Literary Studies (Science Fiction)
• Literary Studies (Travel Literature)
• Literary Studies (War Literature)
• Literary Studies (Women's Writing)
• Literary Theory and Cultural Studies
• Mythology and Folklore
• Shakespeare Studies and Criticism
• Browse content in Media Studies
• Browse content in Music
• Applied Music
• Dance and Music
• Ethics in Music
• Ethnomusicology
• Gender and Sexuality in Music
• Medicine and Music
• Music Cultures
• Music and Media
• Music and Religion
• Music and Culture
• Music Education and Pedagogy
• Music Theory and Analysis
• Musical Scores, Lyrics, and Libretti
• Musical Structures, Styles, and Techniques
• Musicology and Music History
• Performance Practice and Studies
• Race and Ethnicity in Music
• Sound Studies
• Browse content in Performing Arts
• Browse content in Philosophy
• Aesthetics and Philosophy of Art
• Epistemology
• Feminist Philosophy
• History of Western Philosophy
• Meta-Philosophy
• Metaphysics
• Moral Philosophy
• Non-Western Philosophy
• Philosophy of Language
• Philosophy of Mind
• Philosophy of Perception
• Philosophy of Science
• Philosophy of Action
• Philosophy of Law
• Philosophy of Religion
• Philosophy of Mathematics and Logic
• Practical Ethics
• Social and Political Philosophy
• Browse content in Religion
• Biblical Studies
• Christianity
• East Asian Religions
• History of Religion
• Judaism and Jewish Studies
• Qumran Studies
• Religion and Education
• Religion and Health
• Religion and Politics
• Religion and Science
• Religion and Law
• Religion and Art, Literature, and Music
• Religious Studies
• Browse content in Society and Culture
• Cookery, Food, and Drink
• Cultural Studies
• Customs and Traditions
• Ethical Issues and Debates
• Hobbies, Games, Arts and Crafts
• Natural world, Country Life, and Pets
• Popular Beliefs and Controversial Knowledge
• Sports and Outdoor Recreation
• Technology and Society
• Travel and Holiday
• Visual Culture
• Browse content in Law
• Arbitration
• Browse content in Company and Commercial Law
• Commercial Law
• Company Law
• Browse content in Comparative Law
• Systems of Law
• Competition Law
• Browse content in Constitutional and Administrative Law
• Government Powers
• Judicial Review
• Local Government Law
• Military and Defence Law
• Parliamentary and Legislative Practice
• Construction Law
• Contract Law
• Browse content in Criminal Law
• Criminal Procedure
• Criminal Evidence Law
• Sentencing and Punishment
• Employment and Labour Law
• Environment and Energy Law
• Browse content in Financial Law
• Banking Law
• Insolvency Law
• History of Law
• Human Rights and Immigration
• Intellectual Property Law
• Browse content in International Law
• Private International Law and Conflict of Laws
• Public International Law
• IT and Communications Law
• Jurisprudence and Philosophy of Law
• Law and Politics
• Law and Society
• Browse content in Legal System and Practice
• Courts and Procedure
• Legal Skills and Practice
• Legal System - Costs and Funding
• Primary Sources of Law
• Regulation of Legal Profession
• Medical and Healthcare Law
• Browse content in Policing
• Criminal Investigation and Detection
• Police and Security Services
• Police Procedure and Law
• Police Regional Planning
• Browse content in Property Law
• Personal Property Law
• Restitution
• Study and Revision
• Terrorism and National Security Law
• Browse content in Trusts Law
• Wills and Probate or Succession
• Browse content in Medicine and Health
• Browse content in Allied Health Professions
• Arts Therapies
• Clinical Science
• Dietetics and Nutrition
• Occupational Therapy
• Operating Department Practice
• Physiotherapy
• Radiography
• Speech and Language Therapy
• Browse content in Anaesthetics
• General Anaesthesia
• Clinical Neuroscience
• Browse content in Clinical Medicine
• Acute Medicine
• Cardiovascular Medicine
• Clinical Genetics
• Clinical Pharmacology and Therapeutics
• Dermatology
• Endocrinology and Diabetes
• Gastroenterology
• Genito-urinary Medicine
• Geriatric Medicine
• Infectious Diseases
• Medical Toxicology
• Medical Oncology
• Pain Medicine
• Palliative Medicine
• Rehabilitation Medicine
• Respiratory Medicine and Pulmonology
• Rheumatology
• Sleep Medicine
• Sports and Exercise Medicine
• Community Medical Services
• Critical Care
• Emergency Medicine
• Forensic Medicine
• Haematology
• History of Medicine
• Browse content in Medical Skills
• Clinical Skills
• Communication Skills
• Nursing Skills
• Surgical Skills
• Browse content in Medical Dentistry
• Oral and Maxillofacial Surgery
• Paediatric Dentistry
• Restorative Dentistry and Orthodontics
• Surgical Dentistry
• Medical Ethics
• Medical Statistics and Methodology
• Browse content in Neurology
• Clinical Neurophysiology
• Neuropathology
• Nursing Studies
• Browse content in Obstetrics and Gynaecology
• Gynaecology
• Occupational Medicine
• Ophthalmology
• Otolaryngology (ENT)
• Browse content in Paediatrics
• Neonatology
• Browse content in Pathology
• Chemical Pathology
• Clinical Cytogenetics and Molecular Genetics
• Histopathology
• Medical Microbiology and Virology
• Patient Education and Information
• Browse content in Pharmacology
• Psychopharmacology
• Browse content in Popular Health
• Caring for Others
• Complementary and Alternative Medicine
• Self-help and Personal Development
• Browse content in Preclinical Medicine
• Cell Biology
• Molecular Biology and Genetics
• Reproduction, Growth and Development
• Primary Care
• Professional Development in Medicine
• Browse content in Psychiatry
• Addiction Medicine
• Child and Adolescent Psychiatry
• Forensic Psychiatry
• Learning Disabilities
• Old Age Psychiatry
• Psychotherapy
• Browse content in Public Health and Epidemiology
• Epidemiology
• Public Health
• Browse content in Radiology
• Clinical Radiology
• Interventional Radiology
• Nuclear Medicine
• Radiation Oncology
• Reproductive Medicine
• Browse content in Surgery
• Cardiothoracic Surgery
• Gastro-intestinal and Colorectal Surgery
• General Surgery
• Neurosurgery
• Paediatric Surgery
• Peri-operative Care
• Plastic and Reconstructive Surgery
• Surgical Oncology
• Transplant Surgery
• Trauma and Orthopaedic Surgery
• Vascular Surgery
• Browse content in Science and Mathematics
• Browse content in Biological Sciences
• Aquatic Biology
• Biochemistry
• Bioinformatics and Computational Biology
• Developmental Biology
• Ecology and Conservation
• Evolutionary Biology
• Genetics and Genomics
• Microbiology
• Molecular and Cell Biology
• Natural History
• Plant Sciences and Forestry
• Research Methods in Life Sciences
• Structural Biology
• Systems Biology
• Zoology and Animal Sciences
• Browse content in Chemistry
• Analytical Chemistry
• Computational Chemistry
• Crystallography
• Environmental Chemistry
• Industrial Chemistry
• Inorganic Chemistry
• Materials Chemistry
• Medicinal Chemistry
• Mineralogy and Gems
• Organic Chemistry
• Physical Chemistry
• Polymer Chemistry
• Study and Communication Skills in Chemistry
• Theoretical Chemistry
• Browse content in Computer Science
• Artificial Intelligence
• Computer Architecture and Logic Design
• Game Studies
• Human-Computer Interaction
• Mathematical Theory of Computation
• Programming Languages
• Software Engineering
• Systems Analysis and Design
• Virtual Reality
• Browse content in Computing
• Business Applications
• Computer Security
• Computer Games
• Computer Networking and Communications
• Digital Lifestyle
• Graphical and Digital Media Applications
• Operating Systems
• Browse content in Earth Sciences and Geography
• Atmospheric Sciences
• Environmental Geography
• Geology and the Lithosphere
• Maps and Map-making
• Meteorology and Climatology
• Oceanography and Hydrology
• Palaeontology
• Physical Geography and Topography
• Regional Geography
• Soil Science
• Urban Geography
• Browse content in Engineering and Technology
• Agriculture and Farming
• Biological Engineering
• Civil Engineering, Surveying, and Building
• Electronics and Communications Engineering
• Energy Technology
• Engineering (General)
• Environmental Science, Engineering, and Technology
• History of Engineering and Technology
• Mechanical Engineering and Materials
• Technology of Industrial Chemistry
• Transport Technology and Trades
• Browse content in Environmental Science
• Applied Ecology (Environmental Science)
• Conservation of the Environment (Environmental Science)
• Environmental Sustainability
• Environmentalist Thought and Ideology (Environmental Science)
• Management of Land and Natural Resources (Environmental Science)
• Natural Disasters (Environmental Science)
• Nuclear Issues (Environmental Science)
• Pollution and Threats to the Environment (Environmental Science)
• Social Impact of Environmental Issues (Environmental Science)
• History of Science and Technology
• Browse content in Materials Science
• Ceramics and Glasses
• Composite Materials
• Metals, Alloying, and Corrosion
• Nanotechnology
• Browse content in Mathematics
• Applied Mathematics
• Biomathematics and Statistics
• History of Mathematics
• Mathematical Education
• Mathematical Finance
• Mathematical Analysis
• Numerical and Computational Mathematics
• Probability and Statistics
• Pure Mathematics
• Browse content in Neuroscience
• Cognition and Behavioural Neuroscience
• Development of the Nervous System
• Disorders of the Nervous System
• History of Neuroscience
• Invertebrate Neurobiology
• Molecular and Cellular Systems
• Neuroendocrinology and Autonomic Nervous System
• Neuroscientific Techniques
• Sensory and Motor Systems
• Browse content in Physics
• Astronomy and Astrophysics
• Atomic, Molecular, and Optical Physics
• Biological and Medical Physics
• Classical Mechanics
• Computational Physics
• Condensed Matter Physics
• Electromagnetism, Optics, and Acoustics
• History of Physics
• Mathematical and Statistical Physics
• Measurement Science
• Nuclear Physics
• Particles and Fields
• Plasma Physics
• Quantum Physics
• Relativity and Gravitation
• Semiconductor and Mesoscopic Physics
• Browse content in Psychology
• Affective Sciences
• Clinical Psychology
• Cognitive Psychology
• Cognitive Neuroscience
• Criminal and Forensic Psychology
• Developmental Psychology
• Educational Psychology
• Evolutionary Psychology
• Health Psychology
• History and Systems in Psychology
• Music Psychology
• Neuropsychology
• Organizational Psychology
• Psychological Assessment and Testing
• Psychology of Human-Technology Interaction
• Psychology Professional Development and Training
• Research Methods in Psychology
• Social Psychology
• Browse content in Social Sciences
• Browse content in Anthropology
• Anthropology of Religion
• Human Evolution
• Medical Anthropology
• Physical Anthropology
• Regional Anthropology
• Social and Cultural Anthropology
• Theory and Practice of Anthropology
• Browse content in Business and Management
• Business Ethics
• Business Strategy
• Business History
• Business and Technology
• Business and Government
• Business and the Environment
• Comparative Management
• Corporate Governance
• Corporate Social Responsibility
• Entrepreneurship
• Health Management
• Human Resource Management
• Industrial and Employment Relations
• Industry Studies
• Information and Communication Technologies
• International Business
• Knowledge Management
• Management and Management Techniques
• Operations Management
• Organizational Theory and Behaviour
• Pensions and Pension Management
• Public and Nonprofit Management
• Social Issues in Business and Management
• Strategic Management
• Supply Chain Management
• Browse content in Criminology and Criminal Justice
• Criminal Justice
• Criminology
• Forms of Crime
• International and Comparative Criminology
• Youth Violence and Juvenile Justice
• Development Studies
• Browse content in Economics
• Agricultural, Environmental, and Natural Resource Economics
• Asian Economics
• Behavioural Finance
• Behavioural Economics and Neuroeconomics
• Econometrics and Mathematical Economics
• Economic History
• Economic Systems
• Economic Methodology
• Economic Development and Growth
• Financial Markets
• Financial Institutions and Services
• General Economics and Teaching
• Health, Education, and Welfare
• History of Economic Thought
• International Economics
• Labour and Demographic Economics
• Law and Economics
• Macroeconomics and Monetary Economics
• Microeconomics
• Public Economics
• Urban, Rural, and Regional Economics
• Welfare Economics
• Browse content in Education
• Adult Education and Continuous Learning
• Care and Counselling of Students
• Early Childhood and Elementary Education
• Educational Equipment and Technology
• Educational Research Methodology
• Educational Strategies and Policy
• Higher and Further Education
• Organization and Management of Education
• Philosophy and Theory of Education
• Schools Studies
• Secondary Education
• Teaching of a Specific Subject
• Teaching of Specific Groups and Special Educational Needs
• Teaching Skills and Techniques
• Browse content in Environment
• Applied Ecology (Social Science)
• Climate Change
• Conservation of the Environment (Social Science)
• Environmentalist Thought and Ideology (Social Science)
• Management of Land and Natural Resources (Social Science)
• Natural Disasters (Environment)
• Pollution and Threats to the Environment (Social Science)
• Social Impact of Environmental Issues (Social Science)
• Sustainability
• Browse content in Human Geography
• Cultural Geography
• Economic Geography
• Political Geography
• Browse content in Interdisciplinary Studies
• Communication Studies
• Museums, Libraries, and Information Sciences
• Browse content in Politics
• African Politics
• Asian Politics
• Chinese Politics
• Comparative Politics
• Conflict Politics
• Elections and Electoral Studies
• Environmental Politics
• Ethnic Politics
• European Union
• Foreign Policy
• Gender and Politics
• Human Rights and Politics
• Indian Politics
• International Relations
• International Organization (Politics)
• International Political Economy
• Irish Politics
• Latin American Politics
• Middle Eastern Politics
• Political Behaviour
• Political Economy
• Political Institutions
• Political Methodology
• Political Communication
• Political Philosophy
• Political Sociology
• Political Theory
• Politics and Religion
• Politics and Law
• Politics of Development
• Public Policy
• Public Administration
• Qualitative Political Methodology
• Quantitative Political Methodology
• Regional Political Studies
• Russian Politics
• Security Studies
• State and Local Government
• UK Politics
• US Politics
• Browse content in Regional and Area Studies
• African Studies
• Asian Studies
• East Asian Studies
• Japanese Studies
• Latin American Studies
• Middle Eastern Studies
• Native American Studies
• Scottish Studies
• Browse content in Research and Information
• Research Methods
• Browse content in Social Work
• Addictions and Substance Misuse
• Adoption and Fostering
• Care of the Elderly
• Child and Adolescent Social Work
• Couple and Family Social Work
• Direct Practice and Clinical Social Work
• Emergency Services
• Human Behaviour and the Social Environment
• International and Global Issues in Social Work
• Mental and Behavioural Health
• Social Justice and Human Rights
• Social Policy and Advocacy
• Social Work and Crime and Justice
• Social Work Macro Practice
• Social Work Practice Settings
• Social Work Research and Evidence-based Practice
• Welfare and Benefit Systems
• Browse content in Sociology
• Childhood Studies
• Community Development
• Comparative and Historical Sociology
• Disability Studies
• Economic Sociology
• Gender and Sexuality
• Gerontology and Ageing
• Health, Illness, and Medicine
• Marriage and the Family
• Migration Studies
• Occupations, Professions, and Work
• Organizations
• Population and Demography
• Race and Ethnicity
• Social Theory
• Social Movements and Social Change
• Social Research and Statistics
• Social Stratification, Inequality, and Mobility
• Sociology of Religion
• Sociology of Education
• Sport and Leisure
• Urban and Rural Studies
• Browse content in Warfare and Defence
• Defence Strategy, Planning, and Research
• Land Forces and Warfare
• Military Administration
• Military Life and Institutions
• Naval Forces and Warfare
• Other Warfare and Defence Issues
• Peace Studies and Conflict Resolution
• Weapons and Equipment

• < Previous chapter
• Next chapter >

6 Theories of Justification

Richard Fumerton is F. Wendell Miller Professor of Philosophy at the University of Iowa.

• Published: 02 September 2009
• Cite Icon Cite
• Permissions Icon Permissions

The task of explaining and evaluating theories of justification is daunting. There are not only a host of different theories of justification, there are also radical differences among epistemologists concerning how they understand what it is to offer such a theory. This article offers an overview of several prominent positions on the nature of justification. It begins by isolating epistemic justification from nonepistemic justification. It also distinguishes between “having justification for a belief” and “having a justified belief,” arguing that the former is conceptually more fundamental. It then addresses the possibility that justification is a normative matter, suggesting that this possibility has little to offer a concept of epistemic justification. It also critically examines more specific attempts to capture the structure and content of epistemic justification. These include traditional foundationalism and variants thereof, externalist versions of foundationalism; contextualism; coherentism; and “mixed” theories which combine aspects of coherentism and foundationalism.

The concept of justification may be the most fundamental in epistemology. On what became the dominant view in the twentieth century, knowledge is to be understood, at least in part, through our understanding of justification. Part of the answer many offer to Plato's question in the Theaetetus , “What must be added to true belief in order to get knowledge?” is justification. Furthermore, on many accounts of knowledge and justification, it is tempting to conclude that the only responsibility we are competent to carry out qua philosophers is to conform our beliefs to what is justified. Whether or not the world cooperates so as to turn those justified beliefs into knowledge is out of our hands.

The task of explaining and evaluating theories of justification, however, is daunting. There are not only a host of different theories of justification, there are also radical differences among epistemologists concerning how they understand what it is to offer such a theory. Some epistemologists are trying to identify the properties that constitute having justification while others are trying to identify properties upon which justification supervenes. Some philosophers take the product of their analyses to be analytic truths; others claim to be engaged in some sort of empirical investigation. In addition to the fact that there are these meta‐philosophical and methodological controversies lurking in the background, there are serious questions as to whether epistemologists have even agreed on the target of their analyses. Let us begin with this last question.

Epistemic vs. Nonepistemic Justification

The first distinction an epistemologist should emphasize before putting forth a theory of justification is that between epistemic justification and other sorts of justification. If I ask whether S's belief is justified or rational, I might be concerned, for example, with prudential justification. It seems to be a fact that a patient's believing that she will get well often increases the chances of her recovery (even if the resulting probability remains very low). In such a situation there is surely some sense in which the patient would be justified in having (or at least trying to get) the optimistic belief. But even if we allow that there is a sense in which the belief is justified or rational, we don't want to allow that it is epistemically justified or rational. Or consider the person who is becoming paralyzed by fear of death. If believing that there is an afterlife will alleviate that fear and allow the person to live a normal life, then there is again a sense in which it would be perfectly reasonable for that person to try to bring about the belief that has this effect. Prudential reasons for believing (if they exist) have something to do with the efficacy with which believing will or might achieve certain goals or ends.

There may be other nonepistemic reasons for believing or failing to believe a given proposition. It is not wildly implausible to suppose, for example, that a husband has a special moral obligation, and with it a moral reason , to believe that his wife is faithful even in the face of rather powerful epistemic reasons for believing otherwise. One could even imagine a kind of “ 1984 culture” in which one has legal obligations, and legal reasons, to have certain beliefs that are, nevertheless, epistemically irrational.

Can we find a way of characterizing epistemic justification that is relatively neutral with respect to opposing analyses of the concept? As a first stab we might suggest that whatever else epistemic justification for believing some proposition is, it must make probable the truth of the proposition believed. 1 The patient with prudential reasons for believing in a recovery was more likely to get that recovery as a result of her beliefs, but the prudential reasons possessed did not increase the probability of the proposition believed—it was the belief for which the person had prudential reasons that resulted in the increased probability. Epistemic reasons make likely the truth of what is supported by those reasons, and, although it is controversial, it is tempting to suggest that the relation of making likely is not to be understood in causal terms.

Our preliminary characterization of justification as that which makes probable the truth of a proposition may not in the end be all that neutral. As we shall see in a moment, there are those who stress an alleged normative feature of epistemic justification that may call into question the conceptual primacy of probability as a key to distinguishing epistemic reasons from other sorts of reasons. Furthermore, as we shall also see, if one understands the relation of making probable in terms of a frequency conception of probability, one will inevitably beg the question with respect to certain internalist/externalist debates over the nature of justification.

Having Justification for a Belief and Having a Justified Belief

Another preliminary, but important, distinction to stress is that between having justification for a belief and having a justified belief. There seems to be a perfectly clear sense in which there may be enormously strong epistemic reasons for me to believe a given proposition even though I don't end up believing it. In such a situation we can say that there was justification for me to believe the proposition even though I didn't, of course, have a justified belief (or a belief at all) in the relevant proposition. 2 It is tempting to suppose that we can employ the concept of having justification for believing P to define what it is for a person to justifiably believe P. Specifically, one might suggest that a person justifiably believes P when that person believes P and does so based on justification that the person possesses. The analysis of the basing relation is a matter of much controversy. One might hope to analyze it in causal terms. If there is justification J for S to believe P, then S believes P justifiably just in case S's belief is caused by the fact that there is justification for him to believe P. When one presents causal analyses of any concept, however, one should immediately be on guard against counterexamples that rely on “deviant” causal chains. If I possess justification J for believing P, and that causes the hypnotist at the party to hypnotize me into believing P when I hear a doorbell ring, it is not at all clear that I have based the resulting belief on the justification I possess. There may be some relatively straightforward way to revise a causal account of basing to take care of such problems (by, for example, insisting that the causal connection has to be in some sense direct) but we won't explore this issue further here.

If the distinction between possessing justification and having a justified belief is legitimate, which if either of these concepts is more fundamental? If the suggestion made above were plausible, then clearly having justification would be conceptually more fundamental than having a justified belief. We are defining the latter in terms of the former. Furthermore, if we understand the basing relation in causal terms, we should beware of philosophers speculating about which beliefs are or are not justified. One needs empirical evidence to support a causal hypothesis, and it strikes me that philosophers are rarely in possession of the empirical evidence they would need in order to support a psychological claim about what is or is not causing a given belief. Although epistemologists have often supposed that they are trying to determine which beliefs are justified and which are not, I would suggest that if they are to restrict themselves to questions they are competent to answer, qua philosophers, they ought to concern themselves only with the question of whether there is justification for us to believe this or that proposition. Depending on one's analysis of justification, this question may itself end up being an empirical question that philosophers are not particularly competent to address, but this is an issue to which we shall return.

Justification and Normativity

A surprising number of philosophers, with radically different theories of justification, seem to agree that justification is a normative concept. Unfortunately, it is not at all clear what philosophers have in mind by characterizing a concept as normative. We might begin by suggesting that normative terms are those whose meaning can be explicated using paradigm normative expressions, and we might simply list that which is paradigmatically normative. The list might be long or short depending on whether or not we think that all normative expressions can be defined in terms of a relatively few fundamental normative notions. So one might include among the paradigmatically normative such terms as “good,” “ought,” should,” “right,” “permissible,” obligatory,” and their opposites.

If we proceed in this fashion it seems undeniable that the concept of epistemic justification looks suspiciously like a normative concept. As Plantinga ( 1992 ) has effectively reminded us, the etymology of the word “justification” certainly suggests that we are dealing with a value term. And epistemologists often seem quite comfortable interchanging questions about whether or not evidence E justifies one in believing P with questions about whether or not one should believe P on the basis of E. In what is often taken to be an early statement of a justified true belief account of knowledge, Ayer ( 1956 ) described knowledge as true conviction where one has the right to be sure. So again the idea that the concept of justification is normative is at least prima facie plausible. But we must proceed cautiously. We have already seen that we must distinguish epistemic reasons from other sorts of reasons. If we can translate talk about justified belief into talk about what we ought to believe, these same considerations suggest that we must distinguish different senses of “ought.” In the prudential sense of “ought,” perhaps the patient ought to believe she will get better. In the moral sense of “ought,” perhaps the husband ought to believe in his spouse's innocence. But the epistemologist is concerned with what one epistemically ought to believe, and we still need to be convinced that there is some interesting sense in which all of these different “ought” 's express normative concepts.

If we take as our paradigm of a normative “ought” the moral “ought,” then I suppose the question of whether the epistemic “ought” expresses a normative concept reduces to the question of whether there are interesting connections between it and the moral “ought.” The problem now is that moral philosophers have radically different views about what makes moral “ought” judgments normative. Some try to distinguish the normative from the nonnormative by contrasting prescriptive judgements with descriptive judgments. But if this is supposed to be the normative/nonnormative distinction, it is far from clear that the epistemologist should accept the claim that justification is a normative concept. I suspect that most epistemologists take a belief 's being justified to be a fact that admits of description just as straightforwardly as a belief 's having a certain causal history. (Indeed, on some theories of justification a belief 's being justified just is a matter of its causal history.)

Richard Foley ( 1987 ) has suggested that we might understand epistemic justification in terms of what one ought to believe, and he goes on to understand the difference between the epistemic “ought” and other “oughts” 's as differences between species of a common genera. Crudely put, Foley's idea is that normative judgments all assess the efficacy of achieving goals or ends. There are different kinds of normative judgments concerning what we ought to do and what we ought to believe because there are different goals or ends that we are concerned to emphasize. Thus when we are talking about morally justified action, the relevant goal might be something like creating good and avoiding evil. When we are concerned with what prudence dictates, the relevant goals or ends change, perhaps to include everything that is desired intrinsically, for example. What one legally ought to do is a function of the extent to which an action satisfies the goal of following the law. To fit the epistemic “ought” into this framework (and thus classify usefully the kind of normativity that epistemic judgments have) all one needs to do is specify the distinctive goals or ends that define what one epistemically ought to believe. And the obvious candidates are the dual goals of believing what is true and avoiding belief in what is false.

Suggestive as this account might seem, it faces enormous difficulties. It must be immediately qualified to accommodate certain obvious counterexamples. Let's return to our paradigm of a nonepistemic reason, the reason the patient had for believing that she would get well. By forming the relevant belief, the patient might produce for herself a long life which she could devote to scientific and philosophical investigation, investigation that results in an enormous number of true beliefs. Despite accomplishing the goal of believing what is true, our patient (by hypothesis) had no epistemic reason for believing that she would get well. The obvious solution to this problem (one Foley suggests) is to restrict the relevant epistemic goal to that of now believing what is true and now avoiding belief in what is false. But such a revision doesn't really address the problem. Suppose there is an all powerful being who will immediately cause me to believe massive falsehood now unless I accept the epistemically irrational conclusion that there are mermaids. It would seem that to accomplish the goal of believing what is true and avoiding belief in what is false now I must again adopt an epistemically irrational belief.

In desperation one might try restricting the relevant epistemic goal to that of believing what is true now with respect to a given proposition. But now we are in danger of collapsing the distinction between true belief and justified belief. Trivially, the only way to accomplish the goal of believing what is true with respect to P is to believe P when P is true. The problem is that one really wants to identify the content of the epistemic “ought” with what one is justified in believing will accomplish the goal of now believing what is true with respect to a given proposition. But with this revision our “goal” oriented account of epistemic justification becomes pathetically circular. 3

There are, of course, other ways to try to understand the alleged normative character of epistemic justification, but I'm not sure any are illuminating. One might suppose that when one characterizes a belief as justified one is indicating that it is not an appropriate subject for criticism. When one says of a belief that it is unjustified, one is criticizing the belief. For the view to gain even initial credibility, it would be important to distinguish the criticism of a belief from the criticism of the subject who holds the belief. It is simply false that we would always criticize a person for holding a belief we judge to be epistemically irrational. We might, for example, suppose that the person is just too stupid to be able to evaluate properly the relevant evidence and we might, as a result, seldom criticize him for the many wildly irrational beliefs he holds. But even if one makes clear that in characterizing a belief as unjustified one is criticizing the belief not the believer, I'm not sure that one can successfully argue that a person would be guilty of contradiction if, in the grips of some rebellious “anti‐reason” movement, that person criticizes beliefs that conform to the dictates of epistemic rationality.

Once one clearly distinguishes the epistemic “ought” from others it is not in the end clear that one gets much understanding of the concept of justification from the suggestion that epistemic judgements are in some sense normative. 4

Foundationalism

It is tempting to think that one can leave the question of how to understand epistemic justification aside and distinguish different theories of justification in terms of how they understand the structure of epistemic justification. Perhaps the most famous theory of epistemic justification is foundationalism —the very term for the view employs a structural metaphor. But as we shall see, foundationalism is probably best understood not just as a view about the structure of justification. Properly understood, different versions of foundationalism also give an account of the content of epistemic judgements.

Traditional versions of foundationalism have fallen on hard times, but given the present popularity of its externalist cousins, it is still probably the received view in epistemology. Put crudely, the foundationalist believes that all justified beliefs rest ultimately on a foundation of noninferentially justified beliefs. One gets radically different versions of foundationalism depending on how the foundationalist understands noninferential justification.

A little reflection suggests that the vast majority of the propositions for which we have justification have that status only because we justifiably believe other different propositions. So, for example, I justifiably believe that Hitler killed himself, but only because I justifiably believe (among other things) that various generally reliable historical texts describe the event. Foundationalists want to contrast my inferential justification for this belief about Hitler with a kind of justification that is not constituted , in whole or in part, by the having of other justified beliefs. But why should we suppose that there is a kind of justification that is in this way different from inferential justification?

The Regress Arguments for Foundationalism

Suppose I tell you as you approach your fiftieth birthday that you will shortly go insane. I offer as my evidence that you have a genetic defect that, like a time bomb, goes off at the age of 50. Naturally alarmed, you ask me what reason I have for concluding that you have the gene. I respond that it is just a hunch on my part. As soon as you discover that I have no epistemic justification at all for believing that you have the gene, you will immediately conclude that my bizarre conclusion about your impending insanity is wildly irrational. Generalizing from examples like this, one might suggest the following principle:

To be justified in believing P on the basis of E one must be justified in believing E

Now consider another example. Suppose I claim to be justified in believing that Fred will die shortly and offer as my justification that a certain line across his palm (the infamous “lifeline”) is short. Rightly skeptical you wonder this time what reason I have for believing that palm lines have anything whatsoever to do with length of life. As soon as you become satisfied that I have no justification for supposing that there is any kind of probabilistic connection between the character of this line and Fred's life, you will again reject my claim to have a justified belief about Fred's impending demise. 5 That suggests that we might expand our Principle of Inferential Justification (PIJ) to include a second clause:

To have justification for believing P on the basis of E one must not only have (1) justification for believing E, but (2) justification for believing that E makes probable P.

The Epistemic Regress Argument

With PIJ one can present a relatively straightforward epistemic regress argument for foundationalism. If all justification were inferential then for someone S to have justification for believing some proposition P, S must be in a position to legitimately infer it from some other proposition E1. But E1 could justify S in believing P only if S were justified in believing E1, and if all justification were inferential, the only way for S to be justified in believing E1 would be to infer it from some other proposition E2 justifiably believed, a proposition which in turn would have to be inferred from some other proposition E3, which is justifiably believed, and so on, ad infinitum. But finite beings cannot complete an infinitely long chain of reasoning and so, if all justification were inferential, no‐one would be justified in believing anything at all to any extent whatsoever. This most radical of all skepticisms is absurd (it entails that one couldn't even be justified in believing it) and so there must be a kind of justification that is not inferential, that is, there must be noninferentially justified beliefs which terminate regresses of justification.

If we accept the more controversial second clause of PIJ, the looming regresses proliferate. Not only must S above be justified in believing E1, S must also be justified in believing that E1 makes probable P, a proposition that would have to be inferred (if there are no foundations) from some other proposition F1, which would have to be inferred from F2, and so on ad infinitum. But S would also need to be justified in believing that F1 does in fact make likely that E1 makes likely P, a proposition he would need to infer from some other proposition G1, which he would need to infer from some other proposition G2… . And he would need to infer that G1 does indeed make likely that F1 makes likely that E1 makes likely P… . Without noninferential justification, it would seem that we would need to complete an infinite number of infinitely long chains of reasoning in order to be justified in believing anything!

Peter Klein ( 1999 ) has recently defended a view he calls infinitism . The infinitist refuses to accept the existence of noninferential justification, acknowledges that with the availability of only inferential justification, justified belief would require us to be able to come up with infinitely many arguments for infinitely many premises, but argues that finite beings might very well have the capacity to do just that. There is nothing absurd in the supposition that people have an infinite number of justified beliefs (most of which are not, of course, conscious at any given time). You believe justifiably that 2>1, that 3>1, that 4>1, and so on, ad infinitum. While you cannot, of course, complete an infinitely long chain of reasoning, you might be such that you could offer an argument for every proposition you believe. And there is nothing absurd about the suggestion that your ability to do just that is necessary for each of your beliefs being justified.

There seems to be something very odd about the idea that I need arguments to support some of my beliefs, for example, the belief that I'm in pain now, or the belief that I exist now. But even if the availability of an infinite number of dispositional beliefs weakens the foundationalists' claim that without noninferential justification we inevitably face skepticism, it's not clear that the infinitist has a rejoinder to a second regress argument for foundationalism.

The Conceptual Regress Argument

The epistemic regress argument discussed above relies on the unacceptability of a vicious epistemic regress. But one might also argue, more fundamentally, that without a concept of noninferential justification, one faces a vicious conceptual regress. What precisely is our understanding of inferential justification? What makes PIJ true (with or without its controversial second clause). It is at least tempting to answer that PIJ is analytic (true by definition). Part of what it means to claim that someone has inferential justification for believing some proposition P is that his justification consists in his ability to infer P from some other proposition E1 that is justifiably believed. But if anything like this is a plausible analysis of the concept of inferential justification, we face a potentially vicious conceptual regress. Our understanding of inferential justification presupposes an understanding of justification. We need to introduce a concept of noninferential justification in terms of which we can then ultimately define inferential justification.

Consider an analogy. Suppose a philosopher introduces the notion of instrumental goodness (something's being good as a means). That philosopher offers the following crude analysis of what it is for something to be instrumentally good: X is instrumentally good when X leads to something Y, which is good. Even if we were to accept this analysis of instrumental goodness, it is clear that we haven't yet located the conceptual source of goodness. Our analysis of instrumental goodness presupposes an understanding of what it is for something to be good and ultimately presupposes an understanding of what it is for something to be intrinsically good. The conceptual regress argument for foundationalism puts forth the thesis that inferential justification stands to noninferential justification as instrumental goodness stands to intrinsic goodness.

Noninferential Justification

If there is a conceptual regress argument for foundationalism, then one hasn't completed one's foundationalist account of epistemic justification until one gives an account of noninferential justification, an account that itself employs no epistemic concepts. Those who continue to insist that epistemic justification is a normative concept, who reject naturalistic accounts of value, and who further claim that fundamental normative concepts cannot be defined, might claim that an account of noninferential justification consists in an identification of the properties of a belief or a believer upon which noninferential justification supervenes (Goldman 1979 ). The term “supervenience” is a piece of philosophical jargon upon which many these days rely. To say that Y supervenes upon X is usually just to claim that there is some sort of necessary connection between X and Y where one can distinguish as many species of supervenience as one can distinguish kinds of necessary connections. In what follows, I'm going to discuss different accounts of noninferential justification in terms of the conditions with which the proponent of the view identifies having noninferential justification. If one is nervous about identity claims one can translate the views into the language of supervenience.

Noninferential Justification as Infallible Belief

Descartes may be the most well‐known foundationalist. Although he almost never talked about justification (his concern was with knowledge), it seems clear that he embraced the idea that there is a way of knowing that does not rely on what we have called inferential justification. On the most natural interpretation of his views, Descartes identified foundational knowledge with infallible belief. Famously, Descartes found his “first” truth in knowledge of his own existence. What distinguished Descartes's belief about his own existence from other beliefs is that the mere fact that he believed that he existed entailed that he did. Shall we understand noninferential justification in Cartesian terms? Shall we say that S's belief that P is noninferentially justified at t when S's believing P at t entails that P is true?

There are relatively few Cartesian foundationalists around these days. The view is plagued with difficulties. As Lehrer ( 1974 ) and others have pointed out, it is far from clear that this concept of infallible belief has much relevance to our fundamental understanding of noninferential justification. Consider just one technical problem. Every necessary truth is trivially entailed by all propositions (P entails Q when it is impossible for P to be true while Q is false, but if it is impossible for Q to be false then it is entailed by everything). So given the above way of understanding infallible belief, all belief in necessary truth would have noninferential justification. But this just seems wrong. If I whimsically believe some proposition whose necessity is far too complicated for me to grasp, it hardly seems plausible to maintain that the belief would have noninferential justification.

Even if we can find a way of solving the above problem, most contemporary epistemologists are convinced that foundational justification restricted to what can be infallibly believed allows far too insubstantial a foundation to support the complex edifice of what we take ourselves to be justified in believing. There may be a few contingent propositions that are trivially entailed by the fact that they are believed—my belief that I exist, that I have beliefs, that I am conscious—but once we get past propositions whose very subject matter encompasses the fact that they are believed, it's hard to come up with uncontroversial examples of infallible beliefs. As Ayer ( 1956 , 19) argued, as long as the belief that P is one state of affairs and P's being the case is an entirely different state of affairs, it's hard to see how it can be impossible for the former to occur without the latter.

Infallible Justification

Rather than try to identify noninferential justification with some intrinsic feature of a belief that renders the belief infallible, one might instead look for a kind of justification that can accompany a belief and eliminate the possibility of error. Let us say that S's belief that P is infallibly justified at t when S's justification for believing P at t contains as a constituent the very truth‐maker for P. But how can the justification for a belief be identified with a state of affairs that includes as a constituent something that makes true the belief?

Some traditional foundationalists have held that beliefs about experiences are justified by the very experiences that are the subject matter of the beliefs. Thus, for example, it might seem initially plausible to suppose that when I am in pain, it is the pain itself that justifies me in believing that I am in pain. On such a view, the noninferential justification I have for believing that I'm in pain—the experience of pain—trivially guarantees the truth of what I believe. But such a view clearly cries out for some further account of what distinguishes the experience of pain from, say, Caesar's assassination. The above foundationalist wants to claim that while the fact that I'm in pain can justify me in believing that I'm in pain, the fact that Caesar was assassinated cannot justify my belief that Caesar was assassinated. But what is the relevant difference between the two facts that makes it implausible to claim that one is a noninferential justifier, while the other is not? It won't do to call attention to the fact that the pain is an experience of mine. My body is undergoing all sorts of changes right now, the vast majority of which don't justify me in believing that they are occurring. So we still need a principled account of what distinguishes those states of mine that can justify beliefs about them from those states of mine that cannot.

It is tempting to suppose that the foundationalist is better off appealing to some special relation that I have to my pain that makes it unnecessary to look to other beliefs in order to justify my belief that I'm in pain. It's not my pain that justifies me in believing that I'm in pain. It is, rather, the fact that I have a kind of access to my pain that no‐one else has that makes my belief noninferentially justified (while others must rely on inference in order to discover that I'm in this state). The sort of access this foundationalist appeals to is not, of course, justified belief. We need an understanding of noninferential justification that does not rely on an understanding of justified belief. Bertrand Russell ( 1959 and 1984 ) contrasted acquaintance with properties and facts with propositional knowledge. Acquaintance is a sui generis relation that a subject bears to certain facts in virtue of which the subject gets a kind of justification for believing the propositions made true by those facts. A slightly more complicated version of the view maintains that one is noninferentially justified in believing a proposition P when one is directly acquainted with not only the fact that P but also with a relation of correspondence between the thought that P and the fact that P (where the correspondence between a thought and a fact is the essence of a thought's being true). Since acquaintance is a relation that requires the existence of its relata, there is a trivial sense in which one can't possess this sort of justification for believing a proposition while the proposition is false. 6

The acquaintance theory might have one interesting advantage over alternative theories in that it has the potential to offer a unified account of noninferential justification. According to most traditional foundationalists, two of the best candidates for noninferentially justified beliefs are empirical beliefs about the current contents of one's mind and a priori beliefs about relatively straightforward necessary truths. On the acquaintance theory, both direct knowledge of necessary truths and direct knowledge of contingent truths about one's current consciousness would have the same source of justification—acquaintance with facts. The difference between the two kinds of knowledge is not so much a difference in the sources of the knowledge but in the contents of the knowledge. The objects of acquaintance in the case of direct knowledge of mental states are states of affairs whose occurrence is not eternal—the objects of acquaintance in the case of direct knowledge of necessary truths are eternal states of affairs.

Objections to Traditional Foundationalism

In one of the most influential arguments against foundationalism, Wilfrid Sellars ( 1963 , 131–132) argued that the idea of foundational justification as something's being “given” to one in consciousness (something's being an object of direct acquaintance) contains irreconcilable tensions. On the one hand, to ensure that something's being given does not involve any other beliefs, proponents of the view want noninferential justification to be untainted by the application of concepts. On the other hand, the whole point of foundationalism is to end a regress of justification, to give us secure foundational justification for the rest of what we justifiably infer from those foundations. But to make sense of inferences from our foundations, we must ensure that what is given to us in consciousness has a truth value . The kind of thing that has a truth value, however, involves the application of concepts. But to apply a concept is to make a judgment about class membership, and to make a judgment about class membership always involves relating the thing about which the judgment is made to other paradigm members of the class. These judgments of relevant similarity will minimally involve beliefs about the past and thus be inferential in character (assuming that we can have no “direct” access to facts about the past).

The above objection obviously relies on a host of controversial presuppositions. In order to deflect the force of the objection, a traditional acquaintance foundationalist will no doubt emphasize the following. Being directly acquainted with a fact is not, by itself, to have a justified belief in some proposition. It is only acquaintance with a fact conjoined with awareness of a thought's corresponding to a fact that constitutes having noninferential justification. There may well be all kinds of creatures who have acquaintance with facts but no justification for believing anything precisely because they lack the capacity to form thoughts. Secondly, the classical foundationalist will, or at least should, reject the suggestion that to apply a concept is to relate the thing to which one applies the concept to other entities that fall under the concept. Such a view simply invites a vicious regress of the sort that the foundationalist is trying so desperately to avoid. After all, my judgement that X is similar to Y itself involves applying the concept of similarity to the pair X/Y. In doing so am I comparing the pair X/Y to other things that are similar to each other? In fact, I can judge something to be pain without having any recollection whatsoever of any other experience that I have had.

The direct acquaintance theorist does presuppose the intelligibility of acquaintance with facts and in doing so presupposes the intelligibility of a world that has “structure” independently of any structure imposed by the mind. Certain radical versions of “antirealism” reject that commitment to a strong “correspondence” conception of truth and with it the intelligibility of a thought/world fit of which we can be directly aware. 7 While there is some plausibility to the claim that there are, in some sense, alternative conceptual frameworks that we can impose on the world, it is surely absurd to suppose that it is even in principle possible for a mind to force a structure on a literally unstructured world. There are indefinitely many ways to sort the books in a library and some are just as useful as others, but there would be no way to begin sorting books were books undifferentiated. If we couldn't take notice of differences in the world with which we are acquainted, it's not clear how we could “choose” conceptual frameworks with which to make sense of our experience.

Laurence BonJour ( 1985 ) raised another highly influential objection to all forms of classical foundationalism (an objection raised before he himself joined the ranks of the traditional foundationalists). The objection presupposed a strong form of what we might call access internalism. Put superficially, the access internalist argues that a feature of a belief or epistemic situation that makes a belief noninferentially justified must be a feature to which we have actual or potential access. Moreover, we must have access to the fact that the feature in question makes probable the truth of what we believe. So suppose some foundationalist offers an account of noninferential justification according to which a belief is noninferentially justified if it has some characteristic X (where X can stand for any sort of property including complex relational properties). BonJour then argues that the mere fact that the belief has X could not, even in principle, justify the believer in holding the belief. The believer would also need access to (justified belief that!) the belief in question has X and that beliefs of this sort (X beliefs) are likely to be true. At least one of these propositions could only be known through inference, and thus the putative noninferential justification is destroyed.

One must be careful in one's commitment to access requirements for justification lest the view become unintelligible. One can hardly expect an epistemologist to concede that any attempt to identify the conditions X that constitute justification will fail unless one supplements the account with conditions referring to actual or potential access to X. It is immediately clear that one couldn't even in principle satisfy this access internalist. If one tries to supplement X with a believer's having access to X, one simply creates a new condition Y (X plus access) which, according to the view, would itself need to be supplemented by the addition of access requirements to Y. But Y plus access (call it Z) will also be insufficient for justification—we will need to add access conditions to Z, and so on, ad infinitum. The most the access internalist could coherently assert is some sort of necessary connection between having justification and having actual or potential access to justification, where the access in question is not constitutive of the justification. But however one qualifies one's access requirements for justification, access internalism seems far too demanding a theory of epistemic justification. It seems to require of epistemic agents the capacity to form ever more complex justified metabeliefs about the justificatory status of beliefs below.

Traditional Foundationalism and Skepticism

The dissatisfaction with traditional foundationalism probably has as much to do with the threat of skepticism as with any more technical problem facing the view. If we understand noninferential justification in terms of infallible belief or acquaintance with a thought/world fit, on most versions of the traditional view there isn't much we are noninferentially justified in believing. If acquaintance is a real relation, it seems implausible, for example, to suppose that one is directly acquainted with facts about the past or the external world. The following sort of argument seems at least initially powerful:

It's possible that we seem to remember having done something X without having actually done it.

The justification we have for believing that we did X when we have a vivid “hallucinatory” memory would be the same as the justification we have for believing that we did X were we to veridically remember doing X.

The justification we have for believing that we did X when we have vivid “false” memory experience is not direct acquaintance with our having done X (acquaintance is a relation that requires the existence of its relata).

The justification we have for believing that we did X when we have a veridical memory experience is not direct acquaintance with our having done X.

An exactly parallel argument is available with respect to justification for believing propositions about the external world. Such justification never gets any better than the “evidence of our senses.” But,

The justification S has for believing some proposition about the physical world when suffering a vivid hallucinatory experience is the same as the justification S has for believing that proposition were S to have a phenomenologically indistinguishable veridical experience.

The justification S has when hallucinating is obviously not direct acquaintance with some feature of the physical world.

The justification S has in veridical experience is not direct acquaintance with some feature of the physical world.

If the above arguments are sound (they are certainly controversial), it is not entirely clear what will be left in the foundations of empirical justification. The classic empiricist view is that we have noninferentially justified empirical beliefs only about present conscious states. But it has been more than a little difficult to figure out how one can legitimately infer the rest of what we think we are justified in believing from such a limited set of premises. The problem is particularly acute if we accept the second clause of the principle of inferential justification. Given that clause, to advance beyond the foundations of justified belief we would inevitably need to employ nondeductive reasoning and, according to PIJ, that would ultimately require us to have noninferential justification for believing propositions describing probability connections between evidence and conclusions. As long as the relation of making probable is not defined in terms of frequency, as long as making probable is construed as a kind of “quasi‐logical relation” analogous to, but different from entailment, it may not be absurd to suppose that one can have noninferential a priori justification for believing that one set of propositions makes probable another. It is, however, an understatement to suggest that the view is problematic. 8

There is another source of dissatisfaction with the classic empiricist's suggestion that we identify noninferentially justified beliefs with beliefs about the character of present experience. Many would argue, on phenomenological grounds, that we rarely consider propositions describing the intrinsic character of experience. In sense experience our thought is almost always directed out of ourselves and on the existence of an external reality. It requires, the argument goes, considerable effort to turn “inward” to focus on appearance rather than external reality. If most people don't even have the beliefs that the traditional view regards as the only candidates for noninferentially justified beliefs, it seems that once again one faces an unpalatable, fairly extreme, skepticism.

Externalist Versions of Foundationalism

Contemporary externalists offer a refreshingly undemanding account of both non‐inferential and inferential justification. Just about all externalists reject the second clause of the principle of inferential justification. Moreover, noninferential justification is often understood in such a way as to allow for the possibility of a much broader foundation. Consider, for example, the best known version of externalism, Goldman's reliabilism (first set forth in Goldman 1979 ).

The fundamental idea behind reliabilism is strikingly simple. Justified beliefs are reliably produced beliefs. Justified beliefs are worth having because justified beliefs are probably true. The view is a version of foundationalism because it allows us to distinguish two importantly different sorts of justified beliefs—those that result from belief‐independent processes and those that result from belief‐dependent processes. The former are beliefs that are produced by “software” of the brain that takes as its “input” stimuli other than beliefs; the latter are beliefs produced by processes that take as their input at least some other beliefs. So, for example, it is possible that we have evolved in such a way that when prompted with certain sensory input, we immediately and unreflectively reach conclusions about external objects. And we may live in a world in which beliefs produced in such a way are usually true. Crude versions of reliabilism will regard such beliefs as noninferentially justified. Many of our beliefs, of course, result, at least in part, from prior beliefs we hold. We deduce and nondeductively infer a host of propositions describing the world in which we live. Again, on the crudest version of reliabilism, these belief‐dependent processes are reliable when the “output” beliefs are usually true provided that the input beliefs are true.

There are a host of questions that a reliabilist must answer in developing the details of the view. In the crude summary provided above we characterized the reliability of a belief‐independent process in terms of the frequency with which its output beliefs are true. But it takes little imagination to construct counterexamples to this naive a version of the view. Temporary paranoia might cause me to form two, and only two, beliefs about the malicious intentions of my friends, both of which happen to be true. But it hardly seems plausible to suppose that this coincidence makes for a 100 percent reliable belief‐forming process. Minimally, the reliabilist will turn to counterfactuals about the frequency with which output beliefs would be true were the process to produce indefinitely many of them. 9

If we settle the question of how to define reliability, we still need to determine whether the relevant concept of reliability should be relativized to circumstances. It seems obvious that a belief‐forming process might be entirely reliable in one environment, quite unreliable in another. Intuitively, even if the process nets us a majority of true beliefs, we don't want to concede that its operation in the “wrong” environment will result in justified beliefs. The obvious solution would be to define noninferential justification for a given believer in a given environment: S is noninferentially justified in believing P in C at t when S's belief that P in C at t is produced by a belief‐independent process that is reliable in C at t.

While perhaps the most influential, reliabilism is only one version of the externalist alternatives to traditional foundationalism. Armstrong ( 1973 ), for example, suggests the closely related view that some beliefs are noninferentially justified (basic) when they register accurately their subject matter the way an effective thermometer registers accurately the temperature. Although he would resent the suggestion that he is offering a theory of epistemic justification at all, Plantinga ( 1993 ) defines a concept of warrant in terms of beliefs produced by a cognitive apparatus that is properly functioning. He has his own distinctive theistic suggestion for how to understand proper function, but invites allies to try to define the notion in naturalistic terms (for example, in terms of evolutionary history). Plantinga's view is also a version of foundationalism because he holds that properly functioning belief‐producing mechanisms need not involve inference from justified belief.

The most striking feature of most versions of externalism is the way in which they open the door to the possibility of a vastly expanded class of noninferentially justified beliefs. According to the reliabilist, for example, it is never impossible for any belief to acquire noninferential justification. No matter what I believe, it is always in principle possible that the belief is produced by a reliable belief‐independent process. There might be a God who unbeknownst to me causes me to believe with complete conviction a host of true propositions and never causes me to believe a proposition that is false. Such divine inspiration would be a paradigm of a reliable belief‐forming process and the resulting beliefs would all be noninferentially justified. According to the reliabilist, whether or not a given belief is justified depends entirely on whether we are fortunate enough to live in a world in which our cognitive mechanisms produce in us beliefs that are largely successful in getting at the truth.

It is tempting to suppose that externalist versions of foundationalism only delay skeptical problems. Many externalists themselves seem to allow that one can legitimately worry that one has justification for believing that first‐level beliefs are justified. But it is another interesting (some would argue odd) feature of views like reliabilism that there really is no greater problem securing second‐level justification than there is for securing first‐level justification. If, for example, beliefs about the past produced by memory result from reliable belief‐independent processes, then the view implies that beliefs about the past are noninferentially justified. But if beliefs about the past are noninferentially justified then I can easily justify my belief that they are justified. All I need to do is remember that certain beliefs about the past turned out to be true when I relied on memory and employ a standard inductive argument to generalize that beliefs produced in this way are usually true. Of course the classic foundationalist will shudder at this shocking indifference to begging the question. They will protest that one cannot use memory to justify one's belief that memory is reliable! But if reliability really is the essence of justification, it's not clear why one can't study memory using memory to get justified beliefs about its reliability. The investigation into which belief‐forming processes are or are not reliable seems more a task for the cognitive psychologist than for the philosopher, but then perhaps this is why some contemporary epistemologists attempt to straddle the boundary between philosophy and empirical science.

Criticisms of Externalist Foundationalism

If classical foundationalism seemed to require too much in order for us to secure justified beliefs, externalist foundationalism strikes many as requiring too little. At least in a philosophical context, we are interested in having justification because we are interested in gaining a certain sort of assurance of truth. If we start to wonder whether our beliefs about external reality accurately represent that reality, it doesn't seem particularly useful to be told that we may have perfectly justified beliefs provided that they are produced in such a way that they usually accurately represent reality! The justification the philosopher seeks must be such that when one possesses it one's philosophical curiosity is satisfied.

If the primary dissatisfaction with externalism is the feeling that the traditional epistemological questions that have so interested philosophers have simply been redefined in such a way as to change the subject, there are also more technical objections that have been raised to the view. Perhaps the most striking involves a variation on a thought experiment used for a different purpose by Descartes. We'll illustrate the objection focusing again on reliabilism, but variations on the theme affect most externalist analyses of epistemic justification.

Consider a possible world (a kind of Matrix world) in which people are consistently and massively deceived with respect to external reality by a very powerful being. It seems intuitively plausible to suppose that the victims of demonic machination would have precisely the same sort of justification we have for believing (falsely as it turns out) what they do about the world around them. But by hypothesis the demon's victims' beliefs result from unreliable processes, while, we may suppose for the argument, our beliefs result from reliable processes. If the justificatory status of the demon‐world beliefs is the same as those of our world, then it just seems wrong to suppose that reliability is the essence of justification. Since his original paper advocating reliabilism, Goldman himself has struggled with how to respond to the intuitive force of this (and related) objections. After flirting (1986) with the idea of identifying the relevant reliability that defines justification as reliability in “normal” worlds (roughly, worlds in which certain fundamental beliefs we have about this world are true—whether or not they are true in the actual world!), Goldman ( 1988 ) eventually acknowledges two quite distinct concepts of justification: strong (defined by a hard‐core reliabilism in which we simply refuse to acknowledge that demon‐world inhabitants have epistemic justification) and weak (a less demanding concept of epistemic justification roughly defined in terms of meeting “community standards”).

Evidential Externalism

If one is convinced by the externalist that the traditional foundationalist has a concept of justification so demanding that it implies the implausible conclusion that the vast majority of our beliefs are unjustified, one might develop a kind of compromise. One might retain traditional foundationalism, replete with the principle of inferential justification, as capturing a kind of ideal epistemic justification that philosophers seek to attain, but which most people (and most philosophers, for that matter) fail to gain. To soften the blow, one might acknowledge a less demanding concept of epistemic justification that one might be able to satisfy through a kind of nonpropositional analogue of inference. Suppose, for example, that many of our beliefs about the external world are caused by the fact that we have had and are having certain sensations (together with a host of justified background beliefs, most of which remain dispositional). Suppose further that we rarely form beliefs about the character of these sensations, have long since forgotten many of the relevant past experiences (that nevertheless still exert their causal influence), and, of course, rarely, if ever, consciously construct some argument for the ordinary beliefs and expectations we constantly form about the world around us. The facts about sensations that causally contribute to our beliefs about the world are also truth‐makers for propositions (whether we entertain the propositions or not) and it might be the case that the conjunction of propositions made true by the causes of our belief, together with the enormous structure of propositions dispositionally believed that form our epistemic “background,” do make probable (via some sort of legitimate reasoning the epistemologist struggles, usually in vain, to uncover) common‐sense, everyday beliefs. Perhaps we can acknowledge a kind of “unreflective” epistemic justification that we might possess provided that our internal states (including dispositional beliefs and noncognitive states like sensation) satisfy the conditions described above.

Susan Haack ( 1993 ) develops a version of this view but takes a very liberal attitude with respect to what proposition we can employ as the propositional counterpart to sensation. She seems to suggest that we can take the relevant proposition describing a sensation to be one that describes it as the sensation usually produced by a certain physical object under certain conditions. Even if the skeptic allows us a less demanding concept of epistemic justification, that skeptic will no doubt balk at the suggestion that we can take evidential connections between propositions formed this way to be the truth‐makers for claims about epistemic justification. One does need criteria for choosing the propositional counterparts of sensations playing their causal role, but if it is facts that are both causes and truth‐makers for propositions, one can identify the relevant evidential proposition that corresponds to a sensation as the one made true by the fact about the sensation that is causally efficacious in producing the belief in whose epistemic status we are interested.

The above account might seem to be only a minor variation on the concept of epistemic justification defined by reliabilism. Whether this is so depends on how one understands evidential connections. If making probable is a kind of quasi‐logical relation holding between propositions (perhaps even holding necessarily) then the concept of unreflective justification sketched above will be able to resolve the demon‐world objection to reliabilism. The internal causes of belief in the demon world are, by hypothesis, the same as the internal causes of belief in “normal” worlds. The evidential propositional counterparts to the sensory states will be the same, and the justificatory status of the resulting beliefs will be the same. Of course, there may (relative to what we know reflectively) be no evidential connections between the propositions that form our justified background beliefs, the propositions made true by sensation, and the propositions that constitute the conclusions of our common‐sense beliefs, but should that be the case, skepticism wins the day both with respect to demanding and undemanding concepts of epistemic justification.

Contextualism

While the evidential externalist I discussed above is prepared to distinguish more and less demanding standards of justification, the contextualist , for example, Annis ( 1978 ), allows for standards to “float” where the requirements for justified belief are determined in part by the context of inquiry. Recent versions of the view are most often accounts of knowledge. So, for example, Lewis ( 1996 ) suggests that S knows that P when S has a true belief that P where S's evidence eliminates all relevant alternatives to P. What makes the view contextualist is that relevancy is determined by context, including such subjective factors as whether or not the believer is taking seriously the possibility of an alternative. The view is supposed to have the virtue of accommodating both common sense and skepticism—knowledge claims in ordinary contexts will remain true, while in philosophical contexts the skeptic is likely to win the day by forcing us to consider (and thus make relevant) various skeptical scenarios. An analogous view about requirements for justification might allow that one only needs justification for believing certain premises crucial to our reaching conclusions when these background beliefs come under challenge. In ordinary contexts where everyone is happy to allow the truth of our premises and the legitimacy of our reasoning, we can get justified beliefs without having to do what would be necessary were these to come under skeptical challenge.

There is, no doubt, a grain of truth in the contextualist's account of our ordinary, everyday assessments of justification and knowledge. In the context of assessing the justification available for accepting a scientific theory, one simply doesn't worry about the justification we have for believing in the existence of a past or an external world. We assume in the context of such a discussion that we have certain knowledge and that certain forms of reasoning are legitimate, and go on to ask whether on these assumptions, we can legitimately infer the truth of the theory in which we are interested. Philosophers themselves often raise certain objections to common sense beliefs in one philosophical context, only to assume the truth of those very beliefs in a different philosophical context. 10 Monks debating some esoteric proposition concerning the details of their theology may well take for granted the reliability of the Old Testament as a source of truth, presumably knowing full well that should they end up debating an atheist they would need to take a quite different approach.

None of this seems to provide any real support for an interesting form of contextualism, either about knowledge or justification. That we will often “bracket” one set of issues in the context of addressing another, that we will often be interested in seeing what follows from a given set of assumptions, setting aside our ability to “satisfy reason” with respect to those assumptions, is perfectly compatible with our recognizing that in the end our reasons for accepting our conclusions are never really any better than our reasons for accepting the host of background assumptions that remain in the background until we decide to focus our attention upon them. 11

Coherentism

Despite the radical differences among the traditional foundationalists and their more recent externalist counterparts, members of both camps typically share a common conception of the foundational structure of epistemic justification—they are common allies in the fight against coherence theories of epistemic justification.

The coherence theorist rejects the foundationalist's conception of justification as linear . Convinced that there is no escape from the “circle of beliefs”, the coherence theorist argues that we must understand the epistemic justification for a belief in terms of the way in which the proposition believed coheres with other propositions believed. We can distinguish pure and impure coherence theories of justification. A pure coherence theory takes the justification of every belief to be a matter of coherence. An impure theory restricts the thesis to a subclass of beliefs. BonJour ( 1985 ), for example, defended a coherence theory of epistemic justification for empirical beliefs only, but there is nothing in principle to prevent a coherence theorist from restricting the theory to an even more narrow subclass of beliefs.

The vast majority of philosophers who support a coherence theory of justification take the relevant beliefs with which a given justified belief must cohere to be those present in a single individual. What justifies S in believing P is that P coheres with some set of propositions that S occurrently or dispositionally believes (or would believe were S to reflect in a certain way). What justifies you in believing P is P's coherence with other propositions you do or would believe. But while epistemic justification relativized to an individual's belief system is the norm for coherence theories, one finds at least some interest in what we might call a social coherence theory. Roughly, the idea is that what justifies S in believing P is a matter not just of what S believes, but of what others in the community believe. A very crude social coherence theory of epistemic justification might hold that S is justified in believing P only if P coheres with the propositions believed by all or most members of S's community. Because one can distinguish as many different communities as one likes, epistemic justification on this view must always be relativized to a given community. For simplicity, we will focus on the kind of coherence theories that relativize epistemic justification to an individual's belief system, but most of what we say will apply mutatis mutandis to other versions of the view.

Once we are clear about which beliefs a given belief must cohere with in order to be epistemically justified, we'll need more information from the coherence theorist about what constitutes coherence. Often the coherence theorist will begin by claiming that coherence must minimally involve logical consistency, but go on to concede that consistency is far too weak a requirement to constitute the mainstay of coherence. One can imagine a person with a thousand beliefs, none of which have anything to do with any of the others but where each proposition believed is consistent with the conjunction of the others. Such a belief system hardly seems a paradigm of coherence, and we would be reluctant to concede that each has epistemic justification.

In an interesting argument, Foley ( 1979 ) has argued persuasively that consistency among the propositions one believes is not even a necessary condition for the beliefs' being justified. Focusing on lottery‐type situations, Foley argues that we can easily think of a set of inconsistent beliefs each of which is perfectly justified. If there are a thousand people in a lottery that I know to be fair, I can justifiably believe of each participant that he or she will lose and also justifiably believe that not all of them will lose. None of these beliefs is logically consistent with the conjunction of the rest, but each is justified. So the coherence theorist is wrong to tell us that a belief of ours is epistemically justified only if it is consistent with the rest of what we believe. A closely related problem concerns the possibility of admitting into one's belief system a necessary falsehood F. If one believes even one necessary falsehood, then none of one's beliefs will be consistent with the rest of what one believes; the conjunction of a necessary falsehood with any other proposition is itself a necessary falsehood. It seems more than a little harsh, however, to let one philosophical or mathematical error of this sort destroy the possibility of there being any epistemic justification for believing any proposition.

Coherence theorists are wary of requiring too much for the coherence of a belief system. So, for example, one might initially suppose that a model of a coherent belief system might be one in which each proposition believed is entailed by the conjunction of the rest. But one might also worry that such a requirement is far too difficult to come by. In one sense, however, the worry is misplaced. It is actually extremely easy to satisfy the requirement. Indeed, if we include dispositional beliefs, I can confidently claim to have a belief system in which each of my beliefs is entailed by the rest of what I believe. And the same is, or should be, true of everyone who has taken and remembers a course in elementary logic. One of the truth‐functional connectives we all learned was material implication. As long as we know its truth functional definition, we know that if P is true and Q is true then it is true that P materially implies Q and true that Q materially implies P. Consequently, I assume that if we believe P and believe Q, we will also believe (at least dispositionally) that P materially implies Q and that Q materially implies P. But then for any two propositions P and Q that I happen to believe, there will be in my belief system propositions entailing each. P will be entailed by (Q and Q materially implies P) and Q will be entailed by (P and P materially implies Q). The coherence theorist will no doubt be tempted to reply that the belief in the conditionals is entirely parasitic upon the prior beliefs in P and Q, but once one abandons a linear conception of justification, it's not clear what sort of epistemic priority P and Q are supposed to have just because they may have preceded the belief that P is true if and only if Q is true.

Ironically, perhaps probabilistic connections provide a stronger “glue” for coherence than logical relations. So a coherence theorist is likely to claim that a system of beliefs increases its coherence the more the propositions believed stand in probabilistic connections with each other. Explanatory coherence theorists emphasize the importance of having a belief system in which one maximizes the number of propositions believed where one has within one's belief system propositions that can explain the propositions believed. It's difficult, however, to regard entailment as anything other than the limit of making probable, and if it is too easy to come by a belief system in which each proposition believed is entailed by the rest, it's hard to see how one can avoid the problem by emphasizing probability.

There are enough powerful arguments against coherence theories of justification that one need not turn to more problematic concerns. And some objections to a coherence theory do seem to miss the mark. So, for example, some seem to be concerned with the fact that the coherence theorist embraces a radical relativization of justification. But any plausible account of epistemic justification will acknowledge that one person S can be justified in believing P, while another R is justified in believing not‐P. The traditional foundationalist will no doubt trace the difference between the justificatory status of S and R's beliefs to differences in their memories of past experiences, but it is still the case that radical relativization of justification should be embraced as much by traditional foundationalists as by coherentists.

There is, perhaps, the vague concern that a coherence theory of justification makes one's choice of what to believe far too subjective. I want to know what to believe and the coherence theorist tells me to come up with a coherent set of beliefs. But for every coherent set of propositions I entertain, I can think of another set inconsistent with the first but just as internally coherent. Won't this make the epistemic choice of what to believe implausibly arbitrary? If a theory of justification is to give one guidance, and if one were to somehow start one's deliberations about what to believe with no beliefs at all, then it would seem that the coherence theorist gives one no advice at all concerning what to believe. But we are no doubt simply caused to believe firmly certain propositions, and given that we find ourselves with certain beliefs and are trying to determine whether or not to hold still others, it's not clear that the coherence theorist leaves us with no guidance.

A similar response can be made to those who worry that the coherence theorist cuts us off from the world that makes true or false our beliefs. Nothing in the theory, however, precludes the possibility of our beliefs being caused by features of a belief‐independent world. The epistemological coherence theory holds only that whatever the cause of our beliefs, their epistemic status is a function solely of coherence. 12

Perhaps the most devastating criticism of coherence theories was, ironically, put forth by BonJour in the course of defending the view. Earlier we talked about differences between internalists and externalists. One version of internalism (we might call it inferential internalism) insists that evidential connections between propositions believed does nothing to secure justification for the believer unless the believer has access to the fact that the evidential connections hold. We can then distinguish two radically different versions of coherentism. On one version, a belief is epistemically justified provided that it forms a part of a coherent belief system. On the other, a belief is epistemically justified provided that the believer is aware that (has a justified belief that) the belief coheres with the rest of what is believed. The first version of coherentism seems vulnerable to devastating counterexamples. If a person believes a set of propositions that cohere wonderfully when the person has no way of discovering the inferential connections, in what sense are the beliefs justified? Suppose, for example, that I decide to believe every proposition expressed by the fourth sentence of every paragraph in a very sophisticated physics text. Through a miraculous coincidence the propositions I believe cohere wonderfully. Each is made probable by some conjunction of the others. I, however, have no clue as to what the evidential connections are. Would anyone suppose that my good fortune translates into justification?

If we embrace instead access coherentism, then coherentists face the very regress that traditional foundationalists tried so desperately to avoid. To justifiably believe that our beliefs cohere we would need to know first what we believe and second that the propositions believed stand in the appropriate evidential relations. But as coherentists we have no foundations to fall back on. We can't just give ourselves privileged access to propositions describing our own belief states. Our only access to what we believe is through a coherence we discover between our belief that we have certain beliefs and the rest of what we believe. But to discover this coherence we will once again be forced to discover what we believe, and so on, ad infinitum. An equally vicious regress seems to plague any attempt to discover evidential connections. To justify our belief that a given evidential connection obtains, we would need to discover coherence between our belief that the evidential connection obtains and the rest of what we believe. But discovering that coherence would require that we discover another coherence between our belief about coherence and the rest of what we believe, and so on, ad infinitum.

The basic problem facing access coherence theorists is simple. As pure coherence theorists they have no business giving themselves unproblematic access to any facts about the internal or external world, or the world of logical connections. If there really is a “veil” of belief, then beliefs themselves are hidden from us by metabeliefs, which are hidden from us by metametabeliefs, and so on, ad infinitum. Whenever we attempt to get anything before our consciousness we are led on an endless goose chase toward higher‐ and higher‐level metabeliefs.

Mixed Theories

Susan Haack ( 1993 ), Roderick Chisholm ( 1989 ), Ernest Sosa ( 1991 ), and others have suggested that we don't need to choose between foundationalism and coherentism. We can incorporate elements of both. Haack's crossword puzzle metaphor is perhaps the most vivid illustration of the idea. In a crossword puzzle, we are given an initial clue that may lead us to a tentative conclusion about the correct entry in the puzzle. But it is only when our tentative entry “fits” with the other entries we try that we feel confident that we have the correct solution to the puzzle. According to Haack, experience provides a kind of foundational clue with respect to truth, but coherence (fit) is necessary to raise the level of initial credibility to that of epistemic justification. Sosa allows for a kind of animal knowledge resulting from reliable belief‐forming processes (where reliability is relativized to internal and external circumstances) but insists that it is only when one's belief that one has animal knowledge coheres with the rest of one's beliefs that we can turn animal knowledge into reflective knowledge. Although he is one of the most prominent foundationalists, Chisholm allowed that coherence (concurrence) among propositions believed might be one way to raise the epistemic status of those beliefs (69–71).

Such views obviously need to be evaluated carefully, but it is not clear that any concept of justification purportedly captured by the mixed theory cannot be captured by a more straightforward foundationalism. If we have foundational evidence E1 for P1, foundational evidence E2 for P2, and foundational evidence E3 for P3, then instead of insisting that it is coherence among P1, P2, and P3 that raises the epistemic justification for believing each one, why not simply claim that it is the conjunction of E1, E2, and E3 that constitutes a foundational justification for believing each of P1, P2, and P3, where the conjunction of evidence makes more probable P1, for example, than E1 does alone?

A survey of this sort can at best suggest the rich diversity of views about the nature of epistemic justification and the equally rich diversity of objections those views face. In illustrating many of these views and objections I have painted with a very broad stroke. Moreover, there are a host of interesting variations on the views I did discuss that have been defended by able philosophers one would have liked to mention in a survey of this sort. Painting with a broad stroke can still give one a useful “big” picture, and this is all I hoped to accomplish in the preceding remarks.

See Cohen ( 1984 ) for a defense of the idea that a connection to truth lies at the heart of epistemic justification.

We often speak of having some epistemic justification for believing a proposition in contrast with having justification simpliciter. One can have some epistemic justification for believing P when the justification does not even make P more likely to be true than not—it simply increases the probability of P's being true. In what follows I'll almost always be talking about justification as “all‐things‐considered justification” and will use the term in such a way that one has justification for believing P only if, all things considered, the justification makes P more likely than not to be true.

I have argued elsewhere that our understanding of the “ought” of practical rationality and morality is in fact parasitic upon our understanding of the epistemic ought. It is implausible to understand what it is rational or moral to do in terms of the actual consequences that would result from alternatives. Practical and moral reasons seem to have more to do with what one is epistemically justified in believing about consequences.

For a more detailed discussion of this issue, see Fumerton ( 2001 ).

The example may not be fair. It is far from clear that anyone really accepts as legitimate an argument whose premise describes a lifeline and whose conclusion describes length of life. Such arguments may always be enthymemes. The question then becomes whether it is still plausible to claim that one cannot justifiably accept the conclusion of any argument without justifiably believing that there exists the relevant connection between premise and conclusion. I think one can make the case that it is.

There is, however, nothing to prevent an acquaintance theorist from allowing that one can have noninferential justification for believing P that does not entail P's truth. It may be that one can be noninferentially justified in believing P in virtue of being directly acquainted with a fact very similar to, but ultimately different from the fact that P. For an attempt to develop this view in more detail see Fumerton ( 1985 ).

See, for example, Goodman ( 1978 ) and Putnam ( 1988 ).

One of the earliest attempts to construe probability as a relation that holds necessarily between certain propositions was Keynes ( 1921 ).

Or, one could replace talk of frequency in defining reliability with some other notion. Goldman toys with the idea of understanding reliability in terms of an undefined notion of propensity to produce true beliefs.

David Hume ( 1888 ), for example, attacked relentlessly the legitimacy of inductive reasoning and the rationality of belief in an external world, only to assume both in the context of investigating the subject matter of moral judgments.

For a more detailed discussion of contextualism see Moser 1985 , chap. 2.

There is also a coherence theory of truth that might seem a natural ally of the coherence theory of justification. The problems facing a coherence theory of justification, however, pale in comparison to those facing the coherence theory of truth. See Fumerton ( 2001 ).

Annis, David ( 1978 ). “ A Contextualist Theory of Epistemic Justification. ” American Philosophical Quarterly 15: 213–219.

Google Scholar

Armstrong, David ( 1973 ). Belief, Truth and Knowledge . London: Cambridge University Press.

Google Preview

Ayer, A. J. ( 1956 ). The Problem of Knowledge . Edinburgh: Penguin.

BonJour, Laurence ( 1985 ). The Structure of Empirical Knowledge . Cambridge, Mass.: Harvard University Press.

Cohen, Stewart ( 1984 ). “ Justification and Truth. ” Philosophical Studies 46: 279–296. 10.1007/BF00372907

Chisholm, Roderick ( 1989 ). Theory of Knowledge , 3rd ed. Englewood Cliffs, N.J.: Prentice‐Hall.

Foley, Richard ( 1979 ). “ Justified Inconsistent Beliefs. ” American Philosophical Quarterly 16: 147–158.

——. ( 1987 ). The Theory of Epistemic Rationality . Cambridge: Harvard University Press.

Fumerton, Richard ( 1985 ). Metaphysical and Epistemological Problems of Perception . Lincoln: University of Nebraska Press.

——. ( 2001 ). “Epistemic Justification and Normativity.” In Knowledge, Truth and Obligation: Essays on Epistemic Responsibility and the Ethics of Belief , ed. Matthias Steup. Oxford: Oxford University Press.

——. ( 2002 ). Truth and Correspondence . Boston: Rowman & Littlefield.

Goldman, Alvin ( 1979 ). “What Is Justified Belief?” In Justification and Knowledge , ed. George Pappas , 1–23. Dordrecht: Reidel.

——. ( 1986 ). Epistemology and Cognition . Cambridge: Harvard University Press.

——. ( 1988 ). “Strong and Weak Justification.” In Philosophical Perspectives vol. 2: Epistemology , ed. James Tomberlin , 51–69. Atascadero, Calif.: Ridgeview.

Goodman, Nelson ( 1978 ). Ways of World Making . Indianapolis: Hackett.

Haack, Susan ( 1993 ). Evidence and Inquiry . Oxford: Blackwell.

Hume, David ( 1888 ). A Treatise of Human Nature , ed. L. A. Selby‐Bigge . London: Oxford University Press.

Keynes, John ( 1921 ). A Treatise on Probability . London: Macmillan.

Klein, Peter (1999). “Human Knowledge and the Infinite Regress of Reasons.” In Philosophical Perspectives , vol. 13, ed. James Tomberlin , 297–325.

Lehrer, Keith ( 1974 ). Knowledge . Oxford: Oxford University Press.

Lewis, David ( 1996 ). “ Elusive Knowledge. ” Australian Journal of Philosophy 74: 4.

Moser, Paul ( 1985 ). Empirical Justification . Dordrecht: Reidel.

Plantinga, Alvin ( 1992 ). “Justification in the 20th Century.” In Philosophical Issues , vol. 2: Rationality in Epistemology , ed. Enrique Villanueva , 43–78. Atascadero, Calif.: Ridgeview.

——. ( 1993 ). Warrant and Proper Function . Oxford: Oxford University Press. 10.1093/oxfordhb/0195078640.001.0001

Putnam, Hilary ( 1988 ). Representation and Reality . Cambridge, Mass.: MIT Press.

Russell, Bertrand ( 1959 ). The Problems of Philosophy . Oxford: Oxford University Press.

——. ( 1984 ). Theory of Knowledge: The 1913 Manuscript , ed. E. Eames . London: Allen and Unwin.

Sellars, Wilfrid ( 1963 ). Science, Perception and Reality . London: Routledge and Kegan Paul.

Sosa, Ernest ( 1991 ). Knowledge in Perspective . Cambridge: Cambridge University Press.

• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.