Rule-Following and Intransitive Understanding

  • Kjell S. Johannessen
Part of the The Springer Series on Artificial Intelligence and Society book series (HCS)


In logic and philosophy as well as in linguistics, cognitive psychology and computer science there is a particular conception of the nature of language where the concept of rule has a heavy burden. Three kinds of rule, of syntax, inference and semantics, are considered to be necessary in this logico-positivistic model. The chapter considers the flaws and shortcomings of this view of the nature of language and argues for an alternative approach derived from the later philosophy of Wittgenstein. Wittgenstein started out with a rationalist view of this matter but became increasingly aware of the importance of the application of rules. That made him develop a pragmatic conception insisting that the use of language constitutes its nature as the rules cannot dictate their own application. It is only by acting in certain ways, applying words in certains ways, that we are able to make sense. “If language is to be a means of communication there must be agreement not only in definitions but also in judgment”, is Wittgenstein’s way of putting it. There must, accordingly, be a non-interpretative way of grasping a rule. On the basic level, rules are simply followed. That is the point of saying that practices, the established ways of behaving, give words their meaning. Hence rules get their identity from the practices in which they are embedded. To understand a rule is to master the corresponding practices in which it is “inscribed”. The established practices are not isolated monads; they are essentially interrelated. To understand a sentence means to understand a language. Language in turn relates to a form of life.


Human Language Semantic Rule Philosophical Investigation Pragmatic Conception Impossible Situation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    The reference is to Tractatus logico-philosophicus (transl. Pears DF, McGuinness BF), Routledge and Kegan Paul, London, 1961,4.014 and 4.0141.Google Scholar
  2. Below, I am going to refer quite extensively to central writings from Wittgenstein’s Nachlass that have been edited and published as separate books. I shall use the following more or less conventional abbreviations for his writings: T, Tractatus logico-philosophicus (Blackwell, Oxford, 1922) PI, Philosophical investigations (Blackwell, Oxford, 1953) OC, On certainty (Blackwell, Oxford, 1969). PG, Philosophical grammar (Blackwell, Oxford, 1974) ROC, Remarks on colour (Blackwell, Oxford, 1977) RFM, Remarks on the foundations of mathematics (Blackwell, Oxford, 3rd revised edn 1978 ) C & V, Culture and value, ( Blackwell, Oxford, 1980 )Google Scholar
  3. 3.
    Peter Winch tells us that Wittgenstein once made a remark to this effect in conversation with Rush Rhees. See his article, “The unity of Wittgenstein’s philosophy”, that serves as an introduction to the volume Studies in the philosophy of Wittgenstein (London 1969), which he nimself edited. At pp. 12–13, Winch tries to spell out what might be involved in confusing the method of projection with the lines of projection in this context, as that is far from clear.Google Scholar
  4. 25.
    The question of the identity of rules I have treated more fully in my article “Rule following and tacit knowledge”. AI & Society (1988) 2:287–301.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1990

Authors and Affiliations

  • Kjell S. Johannessen

There are no affiliations available

Personalised recommendations