In the same essay, Thomson also tells the story of the violinist newly connected to you: "You wake up in the morning and find yourself back to back in bed with an unconscious violinist. A famous unconscious violinist. He has been found to have a fatal kidney ailment, and the Society of Music Lovers has canvassed all the available medical records and found that you alone have the right blood type to help. They have therefore kidnapped you, and last night the violinist's circulatory system was plugged into yours, so that your kidneys can be used to extract poisons from his blood as well as your own." In nine months the violinist will have recovered from his ailment and can then be safely unplugged from you. But if you unplug him much earlier, then he will surely die. Does this do it? Do you have the intuition that it would be permissible for your to unplug right away?

The three reasons I gave were meant to establish the intermediate conclusion that the criminal law cannot perfectly anticipate all realistically possible cases, cannot be designed so that it always yields the correct result by declaring guilty all and only those who really are guilty. As my third reason for this I adduced the example of rare but morally valid excuses that should nonetheless not be recognized in law. Here is a concrete case. I was caught by a speed camera going 47 mph in a residential area with a 30 mph speed limit. I was going so fast because I was trying to get back home quickly because I had this sudden fear that, lost in thought, I had somehow turned on the gas dial of my kitchen stove without igniting the gas. I feared that gas was filling my apartment, posing a substantial risk to people living above and below. So I drove faster than legal in the reasonable belief that the risk posed by my speeding was very much smaller than the risk posed by my possibly gas-releasing stove....

The legal question varies from country to country, and with regard to the US there is a case to be made that the kinds of instructions you have in mind are speech and therefore protected by the First Amendment. Morally speaking, the case for considering it permissible to prohibit the publication of such instructions seems overwhelmingly strong. There is no legitimate public interest in having such information widely available, and there is a considerable danger to the public from such wide availability. So the public is better served if such publications are prohibited. Here it is important to formulate the prohibition narrowly and with some precision because a broad or vaguely worded statute can easily be used to criminalize or intimidate legitimate political opposition.

Not necessarily. The fact that the adult in question feels OK with the treatment she receives does not show that her feeling is based on a sound judgment. Not long ago, black citizens of the US were routinely treated as inferior creatures: they were scolded for drinking from the whites-only water fountain, for sitting in the front of the bus, and so on. Some blacks, deeply conditioned to feel inferior in a white society, may have felt OK with such abuse ("oh gosh, yes, I am so sorry, I didn't realize this is the whites-only water fountain"). But this hardly shows that there was no abuse going on when blacks were told off in the ways described.

One straightforward way to see that this is not so starts from the realization that the criminal law cannot perfectly anticipate all realistically possible cases. This is so for at least three reasons. First, human powers of anticipation are limited. Second, a criminal law doing justice to all realistically possible cases would be too complex for citizens and officials to comprehend. Third, such a criminal law would also be impossible to administer fairly because it would give criminals too many opportunities to escape punishment. Example: recognizing a rare, morally valid excuse in the law might be a bad idea if the absence of this excuse is very difficult to prove beyond a reasonable doubt. We can expect, then, that in some cases punishments warranted under even the best-designed and -administered criminal law are excessive. In those cases, at least, mercy would not be unjust. For example, we may pardon an offender because we are convinced that he has a morally valid excuse that the law, for good...

We often do nothing in such situations without recognizing this failure to act as a choice, and perhaps even really without making a choice not to act (see Hilary Bok: Acting without Choosing ). Bad faith is manifested not in doing nothing, but in the failure to acknowledge our responsibility for doing nothing. Whether we make a conscious choice or not: we have a choice, we are fully responsible for what we do and we ought to face up to this responsibility -- or so I understand Camus' point here. Doing nothing in the face of an impending bad/evil is sometimes the right thing to do (for example, when intervening would afford little protection to those under threat relative to the costs and dangers to which it would expose third parties) and sometimes at least permissible (for example, when the dangers and cost of intervention to oneself would be high relative to the protection one might afford to those under threat). If, in La Chute , Clamence would have run a serious risk of dying by jumping...

It is true that you often cannot know the outcome of alternative courses of conduct beforehand. But you can typically assign reasonably accurate probabilities, at least a little time forward. We do this all the time when we make decisions -- between two holiday destinations, perhaps, or about whether to accept a job offer or have a child. Utilitarians ask us to do the same sort of thing but then to evaluate in terms of the happiness of all those affected (including oneself). We are to choose the course of conduct that we have good reason to believe will produce the highest probability-weighted expected happiness. The probability-weighted expected happiness produced by a particular course of conduct (C) is calculated this way: one identifies the possible outcomes of C, evaluates each outcome in terms of happiness, multiplies each outcome's happiness by that outcome's probability (subject to one's having chosen C), then sums the products. You are right that one can do this with tolerable accuracy...

Secondary qualities are properties that a thing appears topossess for certain observers of this thing. On reflection, however, secondaryqualities turn out to be ways in which certain observers are affected by thething in question. Colors are an example. Colors are not genuine propertiesinherent in things but rather ways in which human beings with normal eyesightare affected by certain things they encounter. Secondary qualities are thus to be explainedby reference to both: the object with its “primary” qualities and the perceptual apparatus of a specificobserver of this object. The doctrine of secondary qualities brings with it thethought of the object as it is apart from whatever qualities it merely appearsto possess for certain observers. You can call this the thing it itself, thething considered apart from its merely apparent, observer-dependent properties.So the two expressions you query are not at all synonymous but rathercorrelative: “thing in itself” refers to an object as it is apart...

Immanuel Kant used this word in the titles of his most important books, and Kant's use has greatly influenced the sense this word has in later philosophy. By a critique Kant means a critical examination, which will sometimes criticize or undermine a view or method but sometimes also justify and vindicate it.

Absolutely! Readers of such articles usually number in the dozens or, sometimes, hundreds, and very few readers write to the author. So it's really not a burden and, often, a really enlightening experience. You'll find that, in nearly all cases, you'll get a friendly and grateful reply.