Coherence theory

from Wikipedia, the free encyclopedia

In philosophy, coherence theory is called a theory that makes coherence (a connection) with something else a being , a criterion or - in a weak sense - an indication of a thing.

The concept of coherence (connection) is often vague . Sometimes coherence is incorrectly used to mean consistency ( lack of contradiction ).

In a stricter sense, coherence not only presupposes consistency, but also requires that there be inferential, justification and explanatory relationships between the other sentences (justifications). In this stricter sense there are different degrees of coherence.

Coherence theories of truth

Coherence theories of truth (in the broad sense) see the coherence of a statement with other statements as the truth of a statement, the decisive (or only a supplementary) criterion or an indication of the truth of a statement.

In its most widespread, technical sense, the coherence theory of truth is a truth theory that raises coherence to the criterion of truth.

  • According to this, a statement is true if it is part of a coherent system of statements.

As a rule , it is cited as an opposing position to the correspondence theory , which defines truth as the "correspondence" of knowledge and reality . Correspondence theory is about coherence with reality, while coherence theory is about coherence with other statements.

The theory of coherence is related to the consensus theory of truth from which it was adapted.

The objection to the theory of coherence is that there can be coherence to several contradicting sentence systems.

The statement that the earth revolves around the sun is true insofar as it is related to other statements of the Copernican view of the world without contradiction.

Coherence as an indication of the truth of a statement is uncontroversial.

The coherence theory emerged in the 17th century in rationalism , was represented by Hegel and in (English) idealism as well as partly and temporarily by logical empiricism . Coherence theories often go hand in hand with holistic theses. The holism of Quine leads to a certain convergence with the correspondence theory.

Idealistic metaphysical theories teach that there is no ontological type difference between opinions and their truth-makers , only mental entities . Idealists must therefore obviously reject correspondence theories of truth; for correspondence between opinion and object requires objects on the object side that are not themselves opinion-like. Therefore, idealistic theorists would prefer to join a variant of coherence theories.

The classic formulation of the coherence theory comes from HH Joachim , one of the modern representatives of the coherence theories of truth is Nicholas Rescher (see in detail: Truth ).

Coherence theories of justification ( coherentism )

According to the coherence theory of justification (also: coherentism ) “the justification of a single conviction consists in membership in a system of convictions whose individual convictions are among each other [...] in diverse justification relationships”.

According to this, knowledge in the sense of justified opinion does not only consist of a foundation of non-inferential (empirical) knowledge or justified opinion.

Coherentists teach that opinions are only justified by relations with other opinions. In addition to fundamentalist epistemologies (some opinions are justified independently of relations to other opinions), coherentists reject reliabilistic theories (some opinions are justified by being the product of reliable opinion-forming processes).

Taxonomy of the coherence theories of justification

Coherence theories of justification have been divided into classes using different criteria.

Positive and negative coherence theories

Numerous coherence theorists differentiate between positive and negative coherence theories. Gilbert Harman also calls a negative coherence theories general foundations theory .

Negative coherence theories are called coherence theories, which consider all beliefs to be justified initially (prima facie) until something speaks against them. Harman calls this principle of positive undermining and Erik J. Olsson as negative consolidation .

Closely related to negative theories of coherence is the so-called principle of conservatism , according to which a subject is justified to keep his convictions as long as he has no reasons that speak against it.

One problem with negative coherence theories is that, according to negative coherence theories, astrological and religious beliefs are also justified in the first place. Harman calls this objection an anti-religious objection .

Another objection is the paranoia objection , according to which someone who is paranoid about their beliefs is justified according to a negative theory of coherence.

Positive coherence theories are called coherence theories that assume that no belief is justified as long as nothing speaks for it. A belief is justified if it reduces defects in a belief system or increases the coherence of a belief system. Olsson speaks of positive consolidation .

Arbitrary and communicative coherence theories

The distinction between negative and positive coherence theories is not complete. You can also start with a random assessment as justified and unjustified and correct this assessment whenever something speaks for or against a belief. This position is called arbitrary coherence theory.

The beliefs represented by the parents or a scientific teacher, a textbook, etc., can also be considered to be justified in the beginning, in order to then also correct if something speaks for or against a belief. Such coherence theories are called communicative coherence theories.

Computer simulations have shown that negative, positive, arbitrary and communicative coherence theories lead to the same belief systems in the long run.

Coherence theories with degree of embedding

In some coherence theories, relational coherence, i.e. the property of how well a belief fits into a belief system, is a gradual property. In this case, one speaks of coherence theories with a degree of embedding .

The coherence theories of Harold H. Joachim and Francis Herbert Bradley , when read as coherence theories of justification, are coherence theories with a degree of embedding. More recent coherence theories with a degree of embedding are those of Laurence Bonjour , Paul Thagard , Wang, Daniel Schoch and Wiedemann. In constraint satisfaction theories (Thagard, Wang, Schoch, Wiedemann) this corresponds to the degree that considerations have when the networks have found their equilibrium.

Weighted and unweighted coherence theories

In most coherence theories of justification, the coherence of a set of beliefs is determined by inferences (e.g. explanations). If these inferences are differentiated in their strength, one speaks of weighted coherence theories, otherwise of unweighted coherence theories. Examples of weighted coherence theories are those of Thagard, Bartelborth, Wang, Schoch and Wiedemann.

Uncompromising and moderate coherence theories

Uncompromising coherence theories are coherence theories that are both unweighted and have no degree of embedding. If coherence theories do not have at least one of these two properties, one speaks of moderate coherence theories.

Properties of coherent belief systems

There are several conditions that are required of coherent belief systems. Discussions include a .:

  • Condition of the degree of networking: A belief system is all the more coherent, the more inferential relationships (logical and explanatory relationships) the beliefs are networked.
  • Condition of explanatory power: A belief system is all the more coherent, the better the explanations are that link the beliefs.
  • Inconsistency condition: A belief system is more coherent, the fewer contradictions (logical or probabilistic inconsistencies) occur.
  • Subsystem condition: A belief system is all the more coherent, the fewer subsystems it contains that are relatively poorly interlinked.
  • Anomaly condition: the fewer explanatory anomalies that occur, the more coherent a belief system is
  • Condition of competition: A belief system is more coherent, the fewer competing explanations occur.
  • Stability condition: a belief system is more coherent, the more stable the belief system was in the past.

Coherence and consistency

Often, especially by critics of the coherence theory, coherence and consistency are identified, i. This means that only the consistency condition is required of the conditions mentioned.

Joachim and AC Ewing already pointed out, however, that coherence and consistency must not be confused.

Many coherence theorists who have seen the difference consider consistency to be a necessary condition for coherence; H. every coherent belief system is consistent according to this view, but not every consistent belief system is automatically coherent. In addition to Joachim and Ewing, z. B. Stout and Rescher this position.

BonJour has made consistency a necessary condition for coherence in order not to make the coherence theory too complicated.

Bogen has shown that this assumption is implausible for Newton's investigations into the law of gravitation.

The background to such criticism is that rich and complex theories that contain some inconsistencies are better than other theories that are less rich but consistent.

Proponents of coherence theory who do not consider consistency to be necessary usually assume that a coherent belief system should be as consistent as possible.

Probabilistic inconsistencies are a particular problem. H. Beliefs that are not logically contradicting each other, but whose common validity is very unlikely.


Bradley calls for extensive belief systems to be preferred to smaller ones:

"The highter and wider my structure, and the more that any particular fact or set of facts is implied in that structure, the more certain are the structure and the facts."

This property is known as compensating sensitivity. Ewing describes the lack of compensation as

"[...] the fact that such a system of coherent propositions always covers only a very limited part or aspect of reality [...]"

Rescher differentiates between two forms of compensating sensitivity, external compensating sensitivity and internal compensating sensitivity . While the external compensation sensitivity according to Rescher is about including the data as abundantly as possible, the internal compensation sensitivity is about maximizing the system.

In current coherence theories, especially in coherence theories that follow the constraint approach of Thagard and Verbeurgt, but also, for example, in the determinations of coherence by BonJour and Bartelborth , the compensensivity is a derived property that follows from other properties and is not explicitly required.


The different coherence theorists deal very differently with what it means that a belief system breaks down into several independent parts.

Blanshard has demanded, whereby it should be noted that with him causal links also count to the entailments:

"Fully coherent knowledge would be knowledge in which every judgment entailed, and was entailed by, the rest of the system."

Ewing formulated a weaker condition than Blanshard and demanded that every proposition in a completely coherent system can be derived from the rest of the system.

Similarly, according to Bosanquet, a system A, B, C coheres if C follows from A and B, B from A and C and A from B and C. Rescher describes this property as the requirement of (inferential) redundancy .

The characterization of Ewing and Bosanquet has the property that part of a coherent system does not have to be coherent again itself.

As a further condition for coherence, Ewing mentions that a coherent system is a set whose elements are relevant to one another.

The determination of the congruence by Lewis can be seen as a concretization of this condition , who writes that a set of statements is called congruent if and only if the probability of each of its statements increases, if the others are assumed to be true premises.

Chisholm formulates very similarly in his determination of competition that a set A of propositions is competitive for S if and only if A is a set of three or more propositions, each of which is probable for S through the conjunction of the other.

Hansson / Olsson have called this principle the residual support principle.

Price describes a system as coherent when the truth of each of its propositions makes the truth of the rest of the system more likely.

Coherence theory of understanding of persons

In 1989, Read and Miller drew attention to the possibility of a coherence theory of understanding of persons. Kunda , Thagard, Bartelborth and Scholz dealt with this problem later . The basis of a coherence theory of personal understanding is Davidson's forbearance principle, i. H. the idea that we can only understand other people if we accept the coherence of their beliefs; d. H. when we assume that their beliefs are largely inconsistent and coherent.

Coherence Theory of Decision

Thagard and Millgram viewed decisions as problems of coherence about goals and possible actions. Barnes / Thagard extended the coherence theory of decision to emotional decisions. Hurley also made a contribution to the coherence theory of decision. Teacher drew an analogy between acceptance and preference and thus implicitly advocated a theory of coherence of decision.

Coherence theories of ethical and moral justification

Since ethics is often about moral decisions, ethical theories can be built on the basis of coherence theories of decision.

An example of an ethical theory based on coherence is the theory of justice by John Rawls , the justification of which is based on a "balance of thought" between the parties involved in the (hypothetical) choice of the principles of justice .

Coherence theory of concepts

Firth is the founder of the coherence theory of concepts . He assumes that all terms are linked in some way and that we can only fully understand a term if we have also understood the others. According to Firth, a system of conceptual meanings is coherent when the introduction of a new concept affects the entire conceptual system.

Later, the coherence theory of the terms was expanded mainly by Thagard.

Gregory K. Murphy and L. Medin Douglas emphasize that the question of the coherence of concepts is closely related to the question of why certain objects form a concept and others do not. They reject similarity as a measure of the coherence of concepts, since with a suitable similarity measure all concepts are similar to one another. The decisive factor for them is the coherence of the theory in which the terms are used. For Murphy and Medin Douglas, for example, the term apple-or-prime is not very coherent, since a theory that included this term would not be very coherent.


Coherence theories of truth

  • Peter Baumann: Epistemology. 2nd Edition. Metzler, Stuttgart, Weimar. 2006, pp. 175-177.
  • HH Joachim: The Nature of Truth , Oxford University Press, Oxford, 1906.
  • Donald Davidson: A Coherence Theory of Truth and Knowledge , in: Truth And Interpretation, Perspectives on the Philosophy of Donald Davidson, ed. Ernest LePore, Basil Blackwell, Oxford 1986, pp. 307-319.
  • Nicholas Rescher: The Coherence Theory of Truth , Oxford University Press, Oxford 1973.
  • RCS Walker: The Coherence Theory of Truth: Realism, anti-realism, idealism , Routledge, London - New York 1989.
  • JO Young: Global Anti-realism , Avebury, Aldershot 1995.
  • JO Young: A Defense of the Coherence Theory of Truth , in: The Journal of Philosophical Research (26) 2001, pp. 89-101.

Coherent theories of knowledge

  • Peter Baumann: Epistemology. 2nd Edition. Metzler, Stuttgart, Weimar. 2006, pp. 212-215.
  • L. BonJour: The Structure of Empirical Knowledge . Harvard 1985
  • L. Bovens and S. Hartmann: Bayesian Epistemology . Oxford 2003 (German Bayesian epistemology . Paderborn 2006)
  • Keith Teacher: Knowledge . Oxford 1974.
  • Keith Teacher: Theory of Knowledge . Boulder 2 A. 2000.
  • Chisholm / Swartz (eds.): Empirical Knowledge , Prentice-Hall 1973.
  • H. Kornblith: Beyond Foundationalism and the Coherence Theory , in: H. Kornblith (ed.): Naturalizing Epistemology , Bradford Books, MIT Press 1986.
  • E. Olsson: Against Coherence . Oxford 2005.
  • L. Pojman (ed.): The Theory of Knowledge , Wadsworth 3rd A. 2003 (including Dancy: A Defense of Coherentism )
  • J. Pollock: Contemporary Theories of Knowledge . Totowa: Rowman and Allenheld 1986.

Web links


  1. Similarly before Peter Baumann: Epistemology. 2nd Edition. Metzler, Stuttgart, Weimar. 2006, p. 175.
  2. Ernst Tugendhat, Ursula Wolf, Logisch-semantische Propädeutik (1983), p. 240: “The truth that the verification of every sentence is about is ultimately that of the whole system. The agreement theory would now not only be supplemented by the coherence theory, but so mixed that it would have to be redesigned. "
  3. Peter Baumann: Epistemology. 2nd Edition. Metzler, Stuttgart, Weimar. 2006, p. 213.
  4. ^ John L. Pollock, Joseph Cruz: Contemporary Theories of Knowledge. Lanham et al .: Rowman & Littlefield 1999 (2nd ed.), 70 f.
  5. Keith Teacher: Coherentism. In: J. Dancy, E. Sosa (eds.): A Companion To Epistemology. Oxford / Malden, Ma .: Blackwell 1999, 67-70.
  6. Uwe Wiedemann: Theory of Epistemic Justification. Leipzig: Univ. Leipzig 2004, 33 f.
  7. Gilbert Harman: General Foundation versus Rational Insight (Draft September 8, 1999)
  8. ^ A b c Gilbert Harman: Skepticism and Foundations (Draft January 27, 2001).
  9. Gilbert Harman: Positive Versus Negative Undermining in Belief Revision. Nous 18 (1984), p. 46.
  10. ^ A b Erik J. Olsson: Making Beliefs Coherent. The Subtraction and Addition Strategies. Journal of Logic, Language, and Information 7 (1998), p. 144.
  11. a b Uwe Wiedemann: Theory of Epistemic Justification. Leipzig: Univ. Leipzig 2004, p. 34.
  12. Uwe Wiedemann: Theory of Epistemic Justification. Leipzig: Univ. Leipzig 2004.
  13. ^ A b Susan Haack : Evidence and Inquiry: Towards Reconstruction in Epistemology. Oxford / Cambridge, Ma .: Blackwell 1993, p. 18.
  14. a b c Uwe Wiedemann: Theory of epistemic justification. Leipzig: Univ. Leipzig 2004, p. 35.
  15. Laurence BonJour: The Structure of Empirical Knowledge. Cambridge, Mass./London: Harvard University Press 1985, pp. 95-101.
  16. Thomas Bartelborth: Justification strategies. A path through analytical epistemology. Berlin: Akademie Verlag 1996, p. 193.
  17. Uwe Wiedemann: Theory of Epistemic Justification. Leipzig: Univ. Leipzig 2004, p. 61.
  18. Harold H. Joachim: The Nature of Truth. Oxford: Clarendon 1906, pp. 73f.
  19. AC Ewing: Idealism. A critical survey London / New York 1934 (Reprint: 1974), p. 228f.
  20. ^ GF Stout: Immediacy, Mediacy and Coherence. Mind, NS 17 (1908), p. 30f.
  21. Laurence BonJour: The Structure of Empirical Knowledge. Cambridge, Mass./London: Harvard University Press 1985, p. 240 note 7.
  22. James Bogen: Coherentist Theories of Knowledge Don't Apply to enough Outside of Science and Don't Give the Right Results When Applied To Science. In: Bender, John W. (ed.): The Current State of the Coherence Theory. Dordrecht et al: Kluwer Academic Press 1989, p. 148ff.
  23. So z. BFH Bradley: Essays on Truth and Reality. Oxford: Clarendon 1914, p. 202
  24. ^ FH Bradley: Essays on Truth and Reality. Oxford: Clarendon 1914, p. 211.
  25. AC Ewing: Idealism. A critical survey. London / New York 1934 (reprinted 1974), p. 229 footn.
  26. ^ Nicholas Rescher: The Coherence Theory of Truth. Oxford: Clarendon 1973, p. 73.
  27. ^ Brand Blanshard: The Nature of Thought. London: George Allen & Unwin 1939. Vol. II, p. 264.
  28. AC Ewing: Idealism. A critical survey. London / New York 1934 (reprinted 1974), p. 229.
  29. ^ Nicholas Rescher: The Coherence Theory of Truth. Oxford: Clarendon 1973, p. 37.
  30. AC Ewing: Idealism. A critical survey. London / New York 1934 (reprinted 1974), p. 230.
  31. ^ CI Lewis: An Analysis of Knowledge and Valuation. La Selle, Ill. 1946, p. 338.
  32. Chisholm, Roderick, M .: Theory of Knowledge. Englewood, Cliffs .: Prentice-Hall (3rd ed.) 1989, p. 71.
  33. ^ Sven Hansson, Erik J. Olsson: Providing Foundations for Coherentism. Knowledge 51 (1999), p. 246.
  34. ^ HH Price: Perception. London 1932, p. 183.
  35. Stephen J. Read, Lynn C. Miller: Explanatory coherence in understanding persons, interactions, and relationships. Behavioral and Brain Sciences 12 (1989) 3, pp. 485f.
  36. ^ Ziva Kunda / Paul Thagard: Forming Impressions From Stereotypes, Traits, and Behaviors: A Parallel-Constraints-Satisfaction Theory. Psychological Review 103 (1996) 2, 284-308
  37. ^ Paul Thagard: Coherence in Thought and Action. Cambridge / London: MIT Press 2000, xi.
  38. Thomas Bartelborth: Understanding and Coherence. A contribution to the methodology of the social sciences. Analysis and Criticism 21 (1999), 97-116.
  39. ^ Paul Thagard / Elijah Millgram: Inference to the best plan: a coherence theory of decision. In: A. Ram / DB Leake (eds.): Goal-driven learning. Cambridge 1995, pp. 439-454.
  40. ^ Paul Thagard / Elijah Millgram: Deliberative coherence. Synthesis 108 (1996), pp. 63-88.
  41. ^ Allison Barnes / Paul Thagard: Empathy and Analogy. Dialogue 37 (1998), pp. 705-720.
  42. ^ SL Hurley: Natural reasons: personality and polity. New York 1989
  43. Keith Teacher: Self-Trust. A Study of Reason, Knowledge and Autonomy. Oxford: Clarendon 1997, pp. 1ff.
  44. ^ Roderick Firth: Coherence, Certainty, and Epistemic Priority. The Journal of Philosophy 61 (1964), pp. 545-557.
  45. ^ Roderick Firth: Coherence, Certainty, and Epistemic Priority. The Journal of Philosophy 61 (1964), p. 546.
  46. ^ Paul Thagard: Conceptual Revolutions. Princeton, NJ: Princeton University Press 1992.
  47. Gregory K. Murphy / L. Medin Douglas: The Role of Theories in Conceptual Coherence. Psychological Review 92 (1985 (3), 289.
  48. Gregory K. Murphy / L. Medin Douglas: The Role of Theories in Conceptual Coherence. Psychological Review 92 (1985 (3), pp. 291f.
  49. Gregory K. Murphy / L. Medin Douglas: The Role of Theories in Conceptual Coherence. Psychological Review 92 (1985 (3), p. 298.