定義邊界

上文講到analytic philosophy 的一個常用方法。這兩天便讀到了一篇相關的文章。

New APPS: Feynman on precise definitions and philosophical methodology

內裡講到如何辨別某一個定義的邊界，即找出某物的必須與充分條件。但對於這集中處理邊界的做法，作者有以下評論：

Now, one of the problems with this methodology is that it opens the door to a cottage industry of counter-examples, one more far-fetched than the other (Gettier cases, Barn facades… For the real guide, see here), which either satisfy the definiens but not the definiendum, or the other way round. Now, what bothers me most about this approach is that it forces us to focus almost entirely on borderline, atypical cases of something counting as A or satisfying properties B, C, D…; but if the goal is to attain a better knowledge of property A, looking at the borders rather than at the core does not strike me as the most suitable approach. (This observation is corroborated by psychological studies on categorization, suggesting that we categorize mostly taking something to be a prototypical example of A, and then moving on from that – what is known as ‘prototype theory’.) In fact, I have very general reservations regarding Kantian obsessions with borders and demarcations in analytic philosophy; it does seem to entail a certain degree of the 'paralysis of thought' described by Feynman.

這個算是一個頗「大」的說法。若然犧牲了邊界問題，那麼統一性（其實我係指consistency），最後能否得出一個能解決問題的理論呢？

這個邊界問題在法律也出現過，就係Hart 講的No vehicles in the park，咁救護車係唔係vehicle? 由於呢個情景是牽涉人命問題，也許比較容易解答，但如果一般只以抽象概念進行，是有可能忘記研究某定義的重要性。但如果邊界可以模糊，要幾清楚才算足夠呢？最後這個問題還真是幾 analytic school。