On going on the same way

By Dennis J. Darland

Originally written April 29, 1976

Edited January 16, 2008

“… he who intends to end with the inexplicable had best begin with the inexplicable and then say no more, lest he lay himself open to suspicion. If he begin with the inexplicable, saying no more, then this does not prove his helplessness, for it is, anyway, an explanation in a negative sense; but if he does begin with something else and lands in the inexplicable, then this does surely prove his helplessness.” – Kierkegaard.

The autonomy of grammar

The problem of what counts as going on the same way can occur in many contexts. I [in unpublished material] wrote on this extensively in 1976-1977. I am presenting those ideas here, edited to some extent. In the 1^st manuscript [p.19-22] I discuss Wittgenstein’s PI Section 86:

This exposition requires a careful examination. The language game with a table but no schema, I will call (86A). The language game using one schema, I will call (86B), and etc. The language game in which an indefinite number of schema may be used, I will call (86Inf). A game using the same written symbols. But no table or schema, I will call (86).

Now language game (86) is like the language game in PI 2. In both cases no expressions of the rules enter into the language game. The table used in language game (86A), is one sort of expression of the rules governing the language game (86). (Languages, like English, can contain expressions of rules which need not be consulted whenever an order is carried out.) Is language game (86A) incomplete because it contains no expression of the rule governing the use of the table, such as the schema in language game (86B)? No. There is a rule governing the use of the table, but this rule has no expression in the language game (86A). There is however an expression for this rule in language game (86B), etc. Now what about language game (86Inf)? In this game, for each rule used in an order, there is an expression in the language game of another rule which governs this rule. However no order given will there be for every rule (schema) expressed in the order, another rule (schema) explaining the previous rule. Otherwise an infinite number of schemata would have to be included in the order. So it seems that no order in this language is complete. For any order given, another rule could be added to explain the last one included in that order. But, in fact, the order is perfectly complete, unless it turns out that the additional rule is needed to prevent a misunderstanding [Cf. PI 87]

[then in the second paper I wrote in 1976 pp. 24-29] Rules, used in the practice of a language game, are in perfect order, if they fulfill their purpose. Rules which govern the language game, but which have no role in the practice of the language game, cannot be said to be in order or not to be in order. Confusion arises because in some language games, e.g. English, rules which govern the game actually have expression in the language game and play a role in the game.

So since one sometimes needs to be given the rule in the practice of the game, it may seem that it is necessary to be given the rule whenever the rule applies. However, sometimes the rule applies only as a description of the language game played, not as apart of the language game played. If one insists that in every such case, the rule governing the language game be incorporated into the practice of the language game, the language game will fall apart. For no order in the language game satisfies this requirement. No order can be given in which every schema is explained by another schema. One must, at some point, act on an order, although no explanation of the last schema has been given.[Cf. PI, 217, 211] To do so is not to act without a right. If one cannot act without an explanation at some point, then there is no such thing as explanation at any point.[Cf. RFM (1^st ed) V-33]

In language game (86Inf), the last schema given is always governed by the rule which is expressed by the schema

Could there be a language game like (86Inf) except that the last schema is governed by a different rule, say the one expressed by the schema:

Call this language game (86Inf2).

One seems to run into a problem here. Suppose one gives the schema to explain a schema given in an order. As one might give:

In (86Inf). Except to explain a schema given in an order (in (86Inf2).

Here the schema

Is itself is governed by the rule expressed by the schema

But if the schema

Is given as an explanation is acted on according to the rule expressed by

The result would be a DOUBLE shift (I don’t have a graphic)

Thus if the language game is to be governed by

Then when a last schema in an order is explained by another schema, then this schema must be

Rather than

For if

Is acted on by the rule expressed by

So that the schema which was to be explained will be acted upon according to

So that it seems that, after all, the language game described makes sense.

However we said that this game was governed by the rule expressed by

But what does this mean except that the last schema of an order can be further explained by the schema

And this isn’t the case. Well the last schema is acted on in the way a schema is acted upon in language game (86Inf) if

is given as an explanation. So far as language game (86Inf2) is concerned the schema is acted upon according to the schema

So if one wants to say that the last schema of an order in language game (86Inf) must be acted upon according to the rule expressed by

Then one is only expressing something about being called the schema expressing the rule governing the language game. ---not saying something about how one must respond to the final schema given in an order. This is the autonomy of grammar. Exactly analogous cases can be worked out for other cases of going on the same way.

Maybe the connection of this discussion with going on the same way isn’t obvious. How to follow the order can be viewed as how to follow it the same order same way at different times!