J. van Brakel (1993) Unjustified Coherence.. Psycoloquy: 4(23) Frame Problem (7)

Volume: 4 (next, prev) Issue: 23 (next, prev) Article: 7 (next prev first) Alternate versions: ASCII Summary
PSYCOLOQUY (ISSN 1055-0143) is sponsored by the American Psychological Association (APA).
Psycoloquy 4(23): Unjustified Coherence.

Commentary on Fetzer on van Brakel on Ford & Hayes on the Frame Problem

J. van Brakel
Department of Philosophy
University of Utrecht
P.O. Box 80.126
3508 TC Utrecht (Netherlands)



Fetzer's position exemplifies a case of unjustified coherence. He straight-jackets all words and concepts in decontextualised idiosyncratic definitions. This is of little relevance for real-world problems such as the frame problem.


Frame-problem, artificial intelligence, temporal logics, independent persistence, attention, Hume, dynamic frames, qualification problem.
1. In their preface to Reasoning Agents in a Dynamic World: The Frame Problem, Ford & Hayes (1991) wrote: "Fetzer's paper gave rise to considerable controversy during the editorial process, and he asked for a public debate of some of the issues it raised." This led to Hayes's commentary on Fetzer's paper and a reply of Fetzer to Hayes (Fetzer 1991a, Hayes 1991, Fetzer 1991b). In my review of Ford & Hayes (1991; Hayes 1992) I wrote that I tend to side with Hayes in his exchange with Fetzer (van Brakel 1992). Fetzer (1993) characterises my position as logically inconsistent, semantically invalid, empirically wrong, and hence incoherent. I think his charges are based on very deep misunderstandings of my exposition of the Problem of Complete Description (PCD) and in my reply I will do my best to explain it better and further.

2. Fetzer (1993) starts his commentary with a misquotation. He says: "First, in 1.1, van Brakel asserts that there are three dimensions to the frame problem: [A] Which things change and which don't? [B] How can [A] be represented? and, [C] How can/do we reason about [A]?" I do not say this in 1.1 or anywhere else, but in 1.2 I do say that questions [A] - [C] "constitute the family of frame problems." Fetzer may think that quoting "constitute the family of frame problems" as "there are three dimensions to the frame problem" is an acceptable paraphrase, but it is not. The difference is precisely what the PCD (and the family of frame problems) is about. Fetzer assumes that anything, for example, the frame problem, can be given an unambiguous definition: It HAS three dimensions. The EXACT definition of dimension 1 is ..., etc. The PCD says this is not possible; to indicate this I use the family metaphor.

3. Fetzer continues to say that it is logically inconsistent, on my part, to say:

    [i] the "frame problem is not the problem of induction in disguise" AND

    [ii] [A] "is definitely related to the induction problem."

He claims that holding both [i] and [ii] is inconsistent, because [B] and [C] are related to [A]; hence all three of [A] - [C] are related to the problem of induction; hence the frame problem is the problem of induction in disguise. If my vague use of "relate to" (or of "definitely") caused confusion, I apologise. But the overall structure of relations would seem to be very much the same according to both my account and Fetzer's. The problem of induction is more or less closely related to [A]; they are near-relatives. Although I didn't say that in my review, I am quite prepared to leave open the possibility that one may be the other in disguise (whatever that may mean). But the family of frame problems is bigger than just [A] or the problem of induction; hence the frame problem (or, more precisely, the family of frame problems) is not the problem of induction in disguise. One family member, or a single-parent family, is "related to" the extended family, but it is not the extended family in disguise. That is why I said in 1.2: "Part of the confusion about the frame problem is its family resemblance character." and in 1.5: "the conflict about the relation between the frame problem and the induction problem is more indicative of the nature of the frame problem than a substantial symptom in its own right."

4. Fetzer (1993) agrees that "it is logically impossible to provide a complete description of any single event," but he claims that this does not exclude giving "necessary and sufficient causal conditions for events of specific kinds," and he quotes the example of "lighting of a match." I was well aware of this sort of argument; amongst other things, it is spelled out at length in Fetzer (1991ab). I accordingly wrote in 2.2: "There are of course physical laws which are said to relate events [of a certain kind] in a necessary way. But laws (or other universal statements) cannot be applied to concrete events without the addition of unspecified ceteris paribus conditions." And I quoted Fetzer (in 3.4) as an example of one of those who think "one can break out of it [i.e. the PCD] by using heuristics or a panacea like relevance or salience." Fetzer does not dispute the arguments I give there, nor does he object to being quoted in that context. Hence I take it he agrees that appeal to "maximal specificity" (Fetzer 1991a) and the like "is useless as a contribution to solving the frame problem. The problem is just pushed ahead."

5. Fetzer (1991a) discusses "lighting of a match" as a specific case illustrating the requirement of maximal specificity, noting, correctly, that "the satisfaction of the requirement of maximal specificity means that the system thereby described is a `closed system' in relation to the occurrence of a corresponding outcome." In 3.6 I refer to this as an example of those who "recognise the PCD but assume that it does not exist for specific domains." I also point out that on such "closed system"-views it is assumed that "[i] the choice of primitive terms [to describe the properties of a particular closed system] is straightforward and [ii] descriptions can be carried out independently for each domain." Fetzer does not comment on what I write there, nor on my reference in 4.3 to his views on the priority of the physical over the commonsensical. Hence I have to assume he agrees when I say: "It is because these assumptions seem so 'plausible' or 'intuitively' correct that there doesn't seem to be a PCD. But these assumptions are incorrect (van Brakel 1991)."

6. Fetzer (1993) also writes that I am wrong to suggest that one cannot advance "necessary and sufficient definitional conditions for specific concepts"; he offers as proof "the definition of a `bachelor' as an `unmarried adult male'." I do not know why Fetzer thinks this is a criticism of my position. In 3.4 I wrote: "Of course, it is possible to propose conventional definitions specifying necessary and sufficient conditions for, say, the use of a word, but we cannot step outside these definitions and give definitions for all the words that are used in the definitions (or the metalanguage)." Fetzer does not dispute this statement, nor has he provided any reasons to assume that he is able to give definitions of ALL words used in definitions, such that his definitions are unconditionally accepted by all speakers of English (or all students of the University of Minnesota, for that matter).

7. Still, it may be useful to elaborate a bit on my ideas about definitions of concepts (van Brakel & Geurts 1988, van Brakel 1991). The difference between Fetzer and me is, I surmise, that he is talking about the logician's (or rational philosopher's) paradise and I'm talking about the real world. Fetzer would say that the pope is a bachelor and so, according to Fetzer, is a man who has been living with the same woman for 20 years, has brought up eight children (while his "woman" was out earning the money), never looked at another woman or went to a pub, and so on. I wouldn't say that. Fetzer says a bachelor is an unmarried adult male, but how are "unmarried," "adult," and "male" being defined? Does one marry for the church or for the law (what about Muslim law or eighteenth century Nuu-chah-nulth "law")? Can two males be married? What about a man who has been a bachelor all his life and after his death it is discovered that for humanitarian reasons he once married a refugee, whom he never saw? Were people making a mistake calling this man a typical bachelor?

8. Of course, my appeal to real language (which is what the frame problem has to deal with) is of no relevance to Fetzer. He will appeal to an artificial ideal language: such a "logically perfect language (Begriffsschrift) should satisfy the conditions, that every expression grammatically well constructed as a proper name out of signs already introduced shall in fact designate an object, and no new sign shall be introduced as a proper name without being secured a meaning [ = reference]" (Frege 1892). Instead, I would appeal to Davidson (1986, p. 446): "there is no such thing as a language, not if a language is anything like what many philosophers and linguists have supposed." Of course, I do not claim to have finalised the philosophical issues on this matter in my exposition of the Problem of Complete Description. But it is not sufficient simply to appeal to some worn-out positivistic ideas of the 1930s, as Fetzer does. As Quine writes (1990, p. 56): "I would not seek a scientific rehabilitation of something like the old notion of separate and distinct meanings; that notion is better seen as a stumbling block cleared away."

9. Fetzer's position exemplifies a case of unjustified coherence. He straight-jackets all words and concepts in decontextualised idiosyncratic definitions, allowing free-play with the signifiers of COHERENCE, VALIDITY, and CONSISTENCY. This is of little relevance for real-world problems, however, such as the frame problem.


Davidson, D. (1986) A Nice Derangement of Epitaphs. In Truth and Interpretation (E. LePore, ed.). Oxford: Basil Blackwell. Pp. 433-446.

Fetzer, J.H. (1991a) The Frame Problem: Artificial Intelligence Meets David Hume. In Ford & Hayes (1991) 55-70.

Fetzer, J.H. (1991b) Artificial Intelligence Meets David Hume: A Response to Pat Hayes. In Ford & Hayes (1991) 77-86.

Fetzer, J.H. (1993) Van Brakel's Position Appears to be Incoherent. PSYCOLOQUY 4 (14) frame-problem.4.

Ford, K.M. and P.J. Hayes (eds.) (1991) Reasoning Agents in a Dynamic World: The Frame Problem. Greenwich: JAI Press.

Frege, G. (1892) Ueber Sinn und Bedeutung. Zeitschrift fuer Philosophie und Philosophische Kritik 100: 25-50. Quoted from Translations from the Philosophical Writings of Gottlob Frege (P. Geach and M. Black, eds.). Oxford: Basil Blackwell.

Hayes, P.J. (1991) Commentary on "The Frame Problem: Artificial Intelligence Meets David Hume." In Ford & Hayes (1991) 71-76.

Hayes, P.J. (1992) Summary of "Reasoning Agents in a Dynamic World: The Frame Problem" (Ford & Hayes 1991, Eds.). PSYCOLOQUY 3 (59) frame-problem.1.

Quine, W.V. (1990) The Pursuit of Truth. Cambridge MA: Harvard University Press.

van Brakel, J. (1991) Meaning, Prototypes and the Future of Cognitive Science. Minds and Machines 1: 233-257.

van Brakel, J. (1992) The Complete Description of the Frame Problem, PSYCOLOQUY. 3 (60) frame-problem.2.

van Brakel, J. and J.P.M. Geurts (1988) Pragmatic Identity of Meaning and Metaphor. International Studies in the Philosophy of Science 2: 205-226

Volume: 4 (next, prev) Issue: 23 (next, prev) Article: 7 (next prev first) Alternate versions: ASCII Summary