类型：直觉泵

这个直觉泵大概说了什么？

不同国籍的人在不同的地点，通过不同的语言，不同的途径，却可以得出相同的信息：某某杀了人。

这个直觉泵（尚未turn the knobs）想要说什么？

获取相同的特性，不需要有很多的相似之处。

金句

This example of the four different believers is meant to demonstrate how brains could have little in common while still sharing an “intentional” property: believing “the same thing.”

turn the knobs 后的变化

三文鱼没有助听器，椅子比鞋子大

这些和初始的直觉泵——murder in the trafagar square类型不同，后者获知信息的人很明显的获取了这些信息，而前者，（几乎）没人接触过这么直白的信息。说明大脑要理解意义并不是完全做书记员的工作：只能直接获取表面的信息。

金句

So while it might seem obvious that Boris must have learned about the Frenchman by “simply uploading” the relevant Russian sentence in Pravda into his brain, and then “translating” it into, oh, Brainish, there is nothing obvious about the supposition that Boris’s brain performed a similar clerical job (from what into Brainish?) for the fact about salmon.

鸽子与狗，他们假如也在凶杀现场，即使目击了这一过程，也不能最终并保留xx杀人的信息，他们无法保存这种意向性（intentionality）。

于是，这样的turn the knobs有了一种偏向人类中心主义的结论：使用语言者才能获取意义。

金句

So this intuition pump risks carrying a seriously anthropocentric bias into our exploration of meaning. Words and sentences are exemplary vehicles of meaning, but for animals that don’t use them, the idea that their brains nevertheless use them is at least far-fetched—which doesn’t make it false.

意向性(intentionality)

概念

和某某事件，某某特点有关。丹尼特在这里提出意向性是有原因的：几个人都知道xx杀人了，但是这以共享的特点难以用其他词表达，于是意向性在这里派上了用场。它在这里强调，虽然保存或表达信息的形式不同，但他们都与“xx杀人了”有关（aboutness）。

机器难以识别意向性

丹尼特将当时顶尖的计算机：Watson 与 小孩子对比，强调识别意向性并不容易。

金句

The closest we have come yet to creating a general-purpose meaning-detector is IBM’s Watson, which is much better at sorting by meanings than any earlier artificial intelligence system, but notice that it is not at all simple, and would still (probably) misidentify some candidates for death threats that a child would readily get. Even small children recognize that when one laughing kid yells to another, “So help me, I’ll kill you if you do that again!” this is not really a death threat.