12 ideas
| 10496 | Monothetic categories have fixed defining features, and polythetic categories do not [Ellen] |
| Full Idea: Many categories are 'monothetic' (the defining set of features is always unique), and others are 'polythetic' (single features being neither essential to group membership nor sufficient to allocate an item to a group). | |
| From: Roy Ellen (Anthropological Studies of Classification [1996], p.33) | |
| A reaction: This seems a rather important distinction which hasn't made its way into philosophy, where there is a horrible tendency to oversimplify, with the dream of a neat and unified picture. But see Goodman's 'Imperfect Community' problem (Idea 7957). |
| 10497 | In symbolic classification, the categories are linked to rules [Ellen] |
| Full Idea: Symbolic classification occurs when we use some things as a means of saying something about other things. ..They enhance the significance of some categories, so that categories imply rules and rules imply categories. | |
| From: Roy Ellen (Anthropological Studies of Classification [1996], p.35) | |
| A reaction: I'm afraid the anthropologists seem to have more of interest to say about categories than philosophers do. Though maybe we couldn't do anthropology if philosophers had made us more self-conscious about categories. Teamwork! |
| 10495 | Continuous experience sometimes needs imposition of boundaries to create categories [Ellen] |
| Full Idea: Because parts of our experience of the world are complexly continuous, it is occasionally necessary to impose boundaries to produce categories at all. | |
| From: Roy Ellen (Anthropological Studies of Classification [1996], p.33) | |
| A reaction: I like it. Ellen says that people tend to universally cut nature somewhere around the joints, but we can't cope with large things, so the sea tends to be labelled in sections, even though most of the world's seas are continuous. |
| 490 | Everything happens by reason and necessity [Leucippus] |
| Full Idea: Nothing happens at random; everything happens out of reason and by necessity. | |
| From: Leucippus (fragments/reports [c.435 BCE], B002), quoted by (who?) - where? |
| 10498 | Classification is no longer held to be rooted in social institutions [Ellen] |
| Full Idea: The view that all classification finds its roots in social institutions is now generally considered untenable. | |
| From: Roy Ellen (Anthropological Studies of Classification [1996], p.36) | |
| A reaction: And about time too. Ellen (an anthropologist) inevitably emphasises the complexity of the situation, but endorses the idea that people everywhere largely cut nature at the joints. |
| 22717 | Self-interest can fairly divide a cake; first person cuts, second person chooses [Poundstone] |
| Full Idea: To fairly divide a cake between two children, the first divides it and the second chooses. …Even division is best, as it anticipates the second child will take the largest piece. Fairness is enforced by the children's self-interests. | |
| From: William Poundstone (Prisoner's Dilemma [1992], 03 'Cake') | |
| A reaction: [compressed] This is introduced as the basic principle of game theory. There is an online video of two cats sharing a dish of milk; each one drinks a bit, then pushes the dish to the other one. I'm sure two children could manage that. |
| 22718 | Formal game theory is about maximising or minimising numbers in tables [Poundstone] |
| Full Idea: At the most abstract level, game theory is about tables with numbers in them - numbers that entities are are efficiently acting to maximise or minimise. | |
| From: William Poundstone (Prisoner's Dilemma [1992], 03 'Curve') | |
| A reaction: A brilliant idea. The question is the extent to which real life conforms to the numberical tables. The assumption that everyone is entirely self-seeking is blatantly false. Numbers like money have diminishing marginal utility. |
| 22719 | The minimax theorem says a perfect game of opposed people always has a rational solution [Poundstone] |
| Full Idea: The minimax theorem says that there is always a rational solution to a precisely defined conflict between two people whose interests are completely opposite. | |
| From: William Poundstone (Prisoner's Dilemma [1992], 03 'Minimax') | |
| A reaction: This is Von Neumann's founding theorem of game theory. It concerns maximising minimums, and minimising maximums. Crucially, I would say that it virtually never occurs that two people have completely opposite interests. There is a common good. |
| 22720 | Two prisoners get the best result by being loyal, not by selfish betrayal [Poundstone] |
| Full Idea: Prisoners A and B can support or betray one another. If both support, they each get 1 year in prison. If one betrays, the betrayer gets 0 and the betrayed gets 3. If they both betray they get 2 each. The common good is to support each other. | |
| From: William Poundstone (Prisoner's Dilemma [1992], 06 'Tucker's') | |
| A reaction: [by Albert Tucker, highly compressed] The classic Prisoner's Dilemma. It is artificial, but demonstrates that selfish behaviour gets a bad result (total of four years imprisonment), but the common good gets only two years. Every child should study this! |
| 22721 | The tragedy in prisoner's dilemma is when two 'nice' players misread each other [Poundstone] |
| Full Idea: The tragedy is when two 'nice' players defect because they misread the other's intentions. The puzzle of the prisoner's dilemma is how such good intentions pave the road to hell. | |
| From: William Poundstone (Prisoner's Dilemma [1992], 11 'Howard's') | |
| A reaction: I really wish these simple ideas were better known. They more or less encapsulate the tragedy of the human race, with its inability to prioritise the common good. |
| 22722 | TIT FOR TAT says cooperate at first, then do what the other player does [Poundstone] |
| Full Idea: The successful TIT FOR TAT strategy (for the iterated prisoner's dilemma) says cooperate on the first round, then do whatever the other player did in the previous round. | |
| From: William Poundstone (Prisoner's Dilemma [1992], 12 'TIT') | |
| A reaction: There are also the tougher TWO TITS FOR A TAT, and the more forgiving TIT FOR TWO TATS. The one-for-one seems to be the main winner, and is commonly seen in animal life (apparently). I recommend this to school teachers. |
| 22723 | Do unto others as you would have them do unto you - or else! [Poundstone] |
| Full Idea: TIT FOR TAT threatens 'Do unto others as you would have them do unto you - or else!'. | |
| From: William Poundstone (Prisoner's Dilemma [1992], 12 'TIT') | |
| A reaction: Essentially human happiness arises if we are all nice, but also stand up firmly for ourselves. 'Doormats' (nice all the time) get exploited. TIT FOR TAT is weak, because it doesn't exploit people who don't respond at all. |