Fads and Fallacies in the Social Sciences by Steven Goldberg: Part III

Part I Part II

If we thought the last topic was contentious, and it was, then I’m not sure how to class these items. Merely mentioning them will raise blood pressure to dangerous levels in some. Because of this, I will be as careful as I can in discuss Goldberg’s ideas, making clear where mine are different, in as dry a language as I can manage. All emphases are in the original.

Race Exists

And everybody knows it. Goldberg says in this essay (read it!), “not a single black student of mine fails to find risible the claim that there is no such thing as race. Only the occasional white social science major claims to find this contention sensible.” Some, for noble reasons, surely, deny race because of their well-founded fear that some races will be treated, and are treated, and have been treated, worse than others. Interestingly, though Goldberg doesn’t mention this, the same people who deny race will often be the strongest supporters of Affirmative Action (USA) or Positive Discrimination (UK).

But, while race exists and while there are on-average differences between races, the context of these differences matters. If you are looking to increase the probability of finding a speedy runner, you’d do best to travel to East or West Africa and not by hunting around Canada. But if you do not care about runners, then there is no difference between the Africans and Canadians. If you were interested solely in height, then there is a difference in the peoples of Sweden and Laos: if you do not care about height, then there is no difference.

There is a statistical argument that some use to deny the existence of race, and the association with race and intelligence: I examine these next time, when I collect all the statistical methods used by supporters and detractors of Goldberg.

The Death Penalty Discriminates, But Not Like You Think

I am speaking here of the USA only. Goldberg examines, where the race of the perpetrators and victims are known, the two-by-two matrix of white and black murderers and their victims: that is, white murderer-white victim, white murderer-black victim, black murderer-white victim, black murderer-black victim. “Over half of all murders…are committed by blacks, but over half of all those executed for these crimes are white.” The “discrimination is against white murderers—and in favor of the black murderers.”

This “may well represent a feeling on the part of some judges and juries that the life of the white victim (who is usually the victim of the white murderer) is of greater value than the life of the black victim (who is usually the victim of a black murderer).”

Thus, if the discrimination is eliminated, it would have the effect of increasing the number of blacks sentenced to death.

Incidentally, Goldberg reminds us: “A sneer is not an argument.”

The Death Penalty Deters Murderers—Probably

I am speaking again of the USA only. Goldberg stresses continuously that deterrence “refers to other people besides the murderer”: after all, murderers were not deterred. The appropriate question to ask is: did the death penalty deter those who would have murdered? “There are good theoretical reasons, and suggestive statistical empirical reasons, for believing that it does, but these are far from conclusive.”

“It is often claimed that the death penalty is barbaric or uncivilized and that the penalty is vengeful…” Recall that the definition of “barbaric” is an “ought” and is not subject to empirical test.

If the death penalty does deter some people from murdering—and thereby does render the murder rate lower—then the opponent of the death penalty is in the position of claiming that the society permitting a higher murder rate is less barbaric or uncivilized than is one with a lower murder rate.

“There is no doubt that other factors are more important to the deterrence of murder than is the death penalty. The presence of a father during the boy’s years of development is clearly more important.”

Incidentally, by using “boys”—and in other places—Goldberg shows that men murder at a vastly higher rate than women.

Let’s say that the proponent of the death penalty is incorrect in his belief that the death penalty does deter and we do invoke the death penalty. “All” we have done is execute murderers who should not have been executed…and, undeniable and horribly, a very few people who are innocent.

But now let’s say that the opponent of the death penalty is incorrect in his belief tha the death penalty doesn’t deter and we don’t invoke the death penalty. We will be responsible for the deaths of innocent people—those whose deaths would have been prevented by the deterrent effect of the death penalty.

Almost certainly, the number of innocents executed would be less than the number of innocents murdered if the death penalty deters.


The best evidence (Goldberg explores a range of theories) is that “most homosexuality is not purely heredity” but that

heredity does play a predispositional role; that is, Male A, who lacks the predisposing heredity is unlikely to become homosexual no matter what his environment. Male B, who has the predisposing heredity, is not likely to become homosexual if he encounters some environments, but is likely to if he encounter others…

It is commonly, but incorrectly, claimed that some societies positively sanction—or at least do not negatively sanction—homosexuality and that this demonstrates the nonpathological nature of homosexuality in our society [homosexuals have higher rates of depression and suicide than non-homosexuals; the question is whether this is because of socialization or pathology]. Let us ignore the fact that the claim is untrue; while there are societies that practice berdache, permit adolescent homosexual servitude to adults, and the like, no society sees as equal the type of of ongoing homosexuality of which we speak.

(ber·dache: Among certain Native American peoples, a person, usually a male, who assumes the gender identity, and is granted the social status, of the opposite sex. See Thomas Berger’s Little Big Man for a fascinating fictional account.)

Conservatives…see the cost of an acceptance of homosexuality as self-evident…Those on the Right figure that little is lost if they’re wrong; conservatives have always been real good at enduring other people’s pain.

Liberals…feel insecure in acknowledging…that moral and political positions…are inherently founded on subjective assumptions. This insecurity forces liberals to too often make up evidence, as if such evidence—even if correct—could render objective the inherently subjective moral positions that reflect their socialization and psychological needs.

See also what he says about so-called “victimless” crimes—bribery is such a crime, he reminds us.

Goldberg has left out the fact that homosexuality per se is like race: the state is only interesting when it comes to certain behaviors. For example, in many of the medical studies in which I participated (as statistician), the question of whether a man is a homosexual is not asked; what is asked is whether the man has sex with other men; these people are labeled MSM (the acronym being obvious). MSM are at greater risk for certain diseases than men who do not have sex with men.


PLEASE! Please let us refrain from the usual degradation of argumentation here. Let us confine our comments solely to the logic of Goldberg’s argument. Let us have no “I’m for” or “I’m against” statements.

Neither side wishes to face the fact “that the abortion question is inherently and eternally unanswerable, though it can be settled by force…because the question of whether the fetus is or isn’t a person is a matter of the definition of ‘person.'”

If a fetus is a person, then abortion is murder. If a fetus is not a person, up until the moment of birth, then abortion is not murder. Arguments about when a fetus “develops a brain, becomes sentient, can survive outside the mother…cannot answer the ‘what is a person’ in any less arbitrary way than we can. Science cannot refute even the claim…that a baby becomes a person at age two…” Goldberg is once again reminding us that “is” cannot decide what “ought to be.” Please keep this in mind.

“[A] pro-choice advocate’s invocation of a ‘woman’s right to her own body’ merely unsuccessfully attempts to obfuscate…or unnecessarily confuses the issue…” that it is the definition of “person” that matters. If a fetus is a person, then she has no “right” to murder it. In other words, the question of her “right” is no differently, logically, than the question whether the fetus is a person.

Religion cannot be used except by its offering what is and is not a person—the very point in question—nor can the claim that the state “does not have the right” to declare the fetus a person be recognized unless we also recognize, say, that the state cannot decide if a slave is a person.

Goldberg has taken as his basis the “all or nothing” position. This is statement that either the fetus is a person at conception or at birth only. No discussion is given to the idea that the fetus becomes a person in utero: as was, in various cultures, the case; see, for example, the idea of “the quickening”. This was the time, roughly after the first trimester, in which the fetus was thought to be imbued by its soul. Our culture (in the USA in 2009) leans in this direction as witnessed by the debates over “late-term” abortions.

To be clear, the logic of Goldberg’s position is this: only the definition of what a person is matters in this debate. I take it that all of us have a position on this, but that position is irrelevant to what we want to talk about. The only question we want to attack is: is Goldberg’s logic flawed? (I don’t think it is.)

Next Time We examine the most common statistical arguments used against Goldberg and find that they are nearly always used incorrectly, including by some people who ought to know better. I have ignored patriarchy and will leave it until we review the book Why Men Rule.

Part I Part II


  1. masmit

    In order to justify the last sentence about the death penalty, we’d need to determine not just whether it deters, but how commonly it deters. We might also want to know how often it is wrongfully applied.

    Also, it seems to assume a moral equivalence between a murder that might otherwise not have been committed, and a wrongful execution.

    There is also a problem with making us ‘responsible’ for murders we didn’t commit just because we don’t use the death penalty.

    The rightness or wrongness of the death penalty doesn’t rest solely on it’s deterrent effect or lack of same – one might argue that the state should never have the power of life and death over any of it’s citizens, whatever they may have done, even if the price is a higher murder rate.

  2. turingheuristic

    If an abortion is murder and murder is always unacceptable then abortion is always unacceptable. If murder is not always unacceptable then abortion is not always unacceptable even if it remains murder. I guess then it comes down to the question of how one defines murder. If a society can sanction any kind of murder, does this not leave open circumstances where the murder of a baby can be acceptable?

  3. Briggs


    The evidence is, I gather, that executions of the innocent are pretty rare, while if the death penalty deters, it likely does so at a rate higher than that of those wrongly executed. All speculative, of course; but likely.

    Your other point is true: you are arguing for what “ought” to be, and that it might be less morally defensible to execute the innocent than to prevent the deaths of (however many) innocents who might be murdered. There is a “standard” problem in what is misnamed “experimental psychology” which runs something like this (there are many variants): A train is speeding down the hill where it will certainly crash and kill all aboard unless a switch is thrown which will divert the train to another track, where it will certainly kill a man standing on that track. You are at the switch: what do you do?

    Some will not throw the switch (the analogy between capital punishment is here exact) so as not to be personally responsible for the lone man’s death. Whether this is right or wrong is not a matter of what is, but what ought to be.

  4. Briggs


    Quite right: murder is defined as unlawful killing. So again it’s a definition of what ought to be. However, it is interesting to note that in Gairdner’s The Books of Absolutes every culture has a fairly similar definition of “unlawful killing.”

  5. John

    I can’t say I agree with bribery being a victimless crime. Bribery isn’t a crime at all unless someone is vicitmized by downstream effects. At least I can’t think of a case where this is incorrect.

    Criminal bribery is defined as “the offering, giving, receiving, or soliciting of any item of value to influence the actions of an official or other person in discharge of a public or legal duty”. Meaning either some portion of the public is victimized, or the bribe leads to a separate crime.

    Death penalty – When someone commits murder it is either because they didn’t consider the consequences or they didn’t care about the consequences. Maybe my thinking is unique, but I can’t think of any situation when I might do something if the punishment was 20 years in prison, but wouldn’t do it if the punishment was death.

  6. Bernie

    I am confused as to the argument in the paragraphs on race. Can you clarify Goldberg’s fundamental proposition and, in so doing, define race in terms that are minimally subjective thereby avoiding the Clinton self-definition problem?

    The Kohlberg-like moral reasoning dilemma about the switch is excellent. One common manifestation of it, I believe, is the antipathy towards policemen and the military who use deadly force. (I have received some very judgemental and disapproving looks when I tell people that my son is in the military.) It reminds me of the intriguing and seldom elaborated Jack Nicholson speech in the movie “A Few Good Men”. We also have the logic of the commissars, where the switch would be thrown to keep the train running on schedule! In other words there is a calculus as to what an individual’s life is worth against the general welfare of the masses. In turn we come back to individual proclivities for Utilitarian versus Principled bases for action.

    (I just read the Lost Spy by Alexander Meier. It is worth a look.)

  7. DAV

    “Goldberg has taken as his basis the ‘all or nothing’ position. This is statement that either the fetus is a person at conception or at birth only. No discussion is given to the idea that the fetus becomes a person in utero … To be clear, the logic of Goldberg’s position is this: only the definition of what a person is matters in this debate. … The only question we want to attack is: is Goldberg’s logic flawed? (I don’t think it is.)”

    If personhood is not accepted at conception then some other criteria (other than genetics + life) must be employed. That seems to imply (or at least leave open) the possibility that a peson could become a non-person at some later time — depending, of course, upon the deciding criteria.

    I find it hard to argue personhood starting at any point past conception. The three-month “quickening” idea has no basis in evidence so is therefore an arbitrary date. It could just as easily be 21 years. The viability issue could lead (and probably does) lend itself to “pulling the plug” actions. Question: if I need dialysis to live, would I be considered “non-viable” using a definition similar to one employed by pro-choice advocates? Isn’t the need for dialysis conceptually equivalent to a need for umbilical services?

  8. John

    “That seems to imply (or at least leave open) the possibility that a peson could become a non-person at some later time — depending, of course, upon the deciding criteria”

    This already happens if you think about it. I could take a slice of my kidney and grow it indepentently from my body for years in culture. As soon as that slice was removed, I don’t think anyone would argue that it is still human. It still has genetics and life. It still grows, replicates, etc.

  9. Briggs


    Your argument is not quite right logically. If personhood is not accepted at conception it is only true that either at some other point it is so defined (by whom or by what criterion is not relevant logically) or it is never so defined. A solipsist might insist that he is the only person, for example. What you are attempting is to convince others that the criterion or criteria you have decided are the correct ones, or ought to be. The quickening is of course just as arbitrary as “viability.”

    The point here is that there is no logical proof that one criterion is correct and others false. Whatever is chosen must be chosen with regard to a believe about which evidence does not inform.


    I think, for Goldberg, race is genetically and not phenotypically defined. I’ll examine the statistico-logical argument that the “genetic differences between races are minor” next time.


    When I have the book in hand, I’ll flesh out Goldberg’s argument on this.

  10. stan

    John re: death penalty,

    Regardless of whether you would make the distinction, the evidence shows that others do. Megan McArdle had a great post a few years ago on academics falling for the same fallacy with regard to welfare 50 years ago. Academics completely dismissed the notion that giving women cash for having a baby out of wedlock could possibly lead women to choose to have babies for the money. For the academics, they couldn’t imagine a scenario where they themselves would be so stupid.

    Of course, the world does not consist solely of academics.

  11. DAV


    Wasn’t trying to convince anybody — at least intentionally. It just seems that Goldberg’s logic dictates it. Your right about belief but belief isn’t exactly logical. My personal feeling is that defining personhood at some time other than conception leads to conceptual (*ahem*) problems.


    But then, a single organ is far from a complete person. The idea of completeness seems to be part of of the concept although 100% completeness isn’t necessary. (When does a pile of sand become a dune?) Maybe that allows for first trimester non-personhood as the fetus is becoming complete?

    A side note: This is similar in vein to a number of papers (probably class projects) I read on artificial intelligence exploring various modifications to a human. The papers always used first person narrative in the description of the mods then asked “Am I still me?” I think they were collected into a book entitled “Computers and Thought”.

  12. DAV

    Re Death penalty

    I think that statistically, murder victims are personally known to the murder vs. the two being complete strangers. This implies that most murders are crimes of passion occuring without any careful analysis of consequences which, in turn, indicates that the deterrent value of punishment is likely useless. Accepting that the murder should be removed from society, I think the question becomes for how long and should society accept stewardship of the murder including all necssary upkeep with little return?

  13. Mike B

    To answer your narrowly posed question on abortion, Briggs, I would agree that Goldberg’s logic is sound. However, unless there is a universal definition of personhood, Goldberg’s logic doesn’t advance the debate.

  14. JH

    This reminds me a hypothetical situation that I read somewhere (sorry, I am getting old). I will tell it briefly based on what’s in my fading memory. Sally was kidnapped by an organization called SAM. The life of a beloved SAM member Bill was in danger. SAM members connected Sally’s organs to Bill’s for 9 month till Bill could survive on his own. Does SAM have the right to do this to Sally? How would you consider this from a moral standpoint (ought/ought not)?

    I am not looking for answers to the above questions, only want to bring up another dimension to the abortion issue. Goldberg’s position is that only definition of what a person IS matters. Hmmm… definition of a fully conscious person?! I remember Luis Dias had commented a bit on this a while ago.


  15. Briggs

    Mike B,

    I think it does advance the debate, perhaps not in a major way, because many people are arguing about the wrong thing: bringing up “rights” for instance, distracts from understanding what we’re after is a definition of “person.” Whatever we decide, I doubt that there will ever be universal agreement on the definition.


    Perhaps it is true that most victims are known to their murderers, but this is irrelevant to the question whether the death penalty deters. If obviously does not deter those who kill their (known) victims. But has it deterred others who would have killed (their known victims) because the murderers reasoned that if they got caught they would be killed themselves? The answer to that is, I think, yes.

  16. DAV


    I don’t think it irrelevant. My point is that a murder committed in rage likely doesn’t involve thinking about any consequences. Murders committed during commission of crimes somewhere had to include the thought that deadly force has consequences. The same for murder for “business” reasons (a la the mob).

    Recently where I live someone ran over and killed a cop. The suspect dragged from the vehicle seems to have spent a rather bad night and didn’t make it to sunup. I’m sure (without any proof) that will be in the back of the mind of the next person contemplating escape from arrest and will temper some of escape attempt. This is different, though, than someone in a blind rage lashing out. Note, though, that consequences don’t seem to have had any effect on the outcome the suspect encountered.

    The question then becomes: what’s the actual percentage of each (blind vs. calculating)? I think it impossible to answer by analysis how much contemplation of consequences really occurs as only murderers can appear in the sample. So I guess we’re be stuck with the blind/calculating proxyor some equivalent.

  17. John

    In the first 6 weeks of a pregnancy, the embryo undergoes a sort of quality control. ~25% of all pregnancies and ~95% of embryos with genetic abnormalties ‘fail’ QC and are spontaniously aborted. Many of these embryos would have developed into viable human babies had they been carried by one of the small percentage of women with less stringent QC.

    ie nature already provides for first trimester abortions, with different criteria depending on the mother. Pregnancies with less than favorable expected outcomes are frequently terminated early in development.

    Nature will also spontainiously abort when the mother experiences extreme trauma, in theory to increase the likelyhood of the mother surviving, with nature placing more value to the mother than the unborn.

    At the very least I think this shows human definitions are not only arbitrary, but likely have no meaning in nature. The mahayanan buddists have a saying which goes something like this – Elusive answers are frequently the result of asking invalid questions.

    The best you can really hope for, is to make your case, appeal to as many people as possible, and watch policy change with the percentages of people who believe each side. All the while be aware that all sides may be and probably are wrong in principle.

  18. Raven

    We sanction killing in war or self defence.

    One does not need to declare a fetus a non-person in order to sanction its killing.

    We just have to agree that the killing was justified.

  19. Bernie

    And the justification is? It seems to me that the parallel you try to draw automatically bestows personhood on that which is to be killed. You want to go to the polls with that argument?

  20. Alan Grey

    I agree that a definition of ‘person’ is problematic, as you say…

    “The point here is that there is no logical proof that one criterion is correct and others false. Whatever is chosen must be chosen with regard to a believe about which evidence does not inform.”

    Do we just go towards a cultural or individual moral relativism then, where one group can enforce their conception of personhood on everyone (Just as it was with many instances of slavery)? Or do we instead, operate on the empirical and unequivocal evidence that a human life begins at conception, and abandon a separate notion of ‘personhood’ entirely?

  21. Steven Goldberg’s discussion of race is framed as belief versus disbelief in the concept of race, which he defines as “a group that has been subject to strong enough selective pressures for long enough, with low enough gene flow, to end up demonstrably different from other groups.”

    That definition, on its face, is hardly assailable, but it is also so vague as to be useless. That is, some social demographers “see” five races (and they color code them). But other social demographers claim to discern dozens, hundreds, or even thousands of races. As in other descriptive sciences, there are lumpers and splitters.

    In his essay, Goldberg uses Japanese, Tutsis, Zulus, Norwegians, Kenyans, East Africans, West Africans,and Vietnamese as examples of discernibly different peoples. One must assume then that he is a splitter, and ascribes to rather more races than less. He does not list them and he doesn’t give a number, but using his named races as a hint, we can guess he discerns more than 100 and probably more than 500 races.

    Nor does he mention half-breeds, those children of mixed race parentage, which given his proclivity for racial discernment must number in the hundreds of millions.

    I don’t argue with Goldberg on any of this, but must point out that researchers who use “race” as an explanatory factor are hampered by his abundance of races. If a researcher can color code the race variable into five easy bins, the analysis is simplified, as compared to an analysis which codes the race variable into 100 purebred bins plus 100 x 100 halfbreed bins, not to mention all the fractional and 3-way crosses.

    Goldberg’s point is that the denying the existence of race is a fallacious fad. But his definition of race and his splitter proclivity makes that point moot for all practical purposes. I find that kind of self-defeating argument kind of mushy rather than controversial.

  22. Bernie

    Well and neatly said. It seems to me that as Matt and Goldberg hinted at the number of races and therefore the concept of race is still somewhat elastic and in large measure may be defined by the question one is trying to ask and answer. (I’ll stop here because I sense the ghost of Wittengenstein in the offing!)

  23. Briggs

    Mike, Bernie,

    The summary I gave in terms of runners better summarizes Goldberg’s views on race; he had more room to work with in the book than with the article. The gist: context matters. When a trait is not interesting, differences are unimportant. In the linked paper, Goldberg unfortunately uses shorthand for race, letting us mentally fill in the blanks for traits where the context matters. When speaking informally, this does no harm as we all know what each other is discussing (like in terms of Affirmative Action). Perhaps it would have been better if he opted for the tedious process of being explicit.

    Alan Grey,

    Amen, brother. We have no hope by appealing to logic. Yet I am no relativist; further, I claim nobody is. I’ll discuss this more in the next installment.


    Good logical point. But in order to be a “justified killing” you are assuming you are killing a person. All the rest then follows.

    Of course, some consider killing other creatures besides humans as unjustified, but their definition is, of course, arbitrary.

  24. Bernie

    I have been skimming reviews of Goldberg’s books – I am still, though impatiently, waiting for my copies – and it seems to me that the “is” versus “ought” distinction is recognized by less than half of the reviewers and seldom by his negative reviewers. Some of the negative reviews are quite funny. To paraphrase one of the more coherent 3 star Amazon Reviewers: I think I recall hearing of a few matriarchical tribes in some part of New Guinea, but my professor was not clear so I am kind of fuzzy (sic)!

    Other more positive reviews pointed to some of the reviews that accompanied the initial publishing of the book. Perhaps these would be interesting to amass. I have to go through various surreptious avenues to access JSTOR. You might have easier and more complete access.

  25. Briggs


    Good idea. It’d be great to see the academic reviews of the book.

    Unfortunately, I don’t have access to JSTOR (nor the OED, dammit).

  26. Briggs


    Goldberg answers your counter claim in both books. This quote is from When Wish Replaces Thought.

    There is no a priori reason for assuming that [the death penalty deterrent effect] is less relevant to emotional acts than rational acts; most husbands, when agry, slam doors, shout, or sulk. The fact that the death penalty has not deterred the husband who does kill his wife does not cast doubt on the role of the death penalty in deterring those husbands who don’t.

    He goes into much more detail, of course.

  27. Mike B


    As Briggs, states, a fair logical point that there are justified killings of persons. However the justifications for killing in war and self defence don’t work as justifications for killing a fetus, given that one has accepted it as a person.

  28. Bernie

    My free access to JSTOR is via the Boston Public Library. Perhaps the NYPL provides similar free access. Public libraries are a great mark of civilization and definitely one of Benjamin Franklin’s better ideas. They are one of the few public institutions I voluntarily support – outside of police, fire and public health. I will look and send you via email what I find.

  29. JH

    Why There’s No Point to Arguing About Abortion

    I love the title (and the content) of the Chapter that discusses abortion.

  30. Gary Moran

    “Almost certainly, the number of innocents executed would be less than the number of innocents murdered if the death penalty deters.”

    This argument, and that of Goldberg is, IMO, unconvincing. It equates a person sanctioned by the state of killing criminals and probably some innocent people; against the greater number of probable deaths of innocents killed by others. These are patently not equitable, regardless of the deterrent value.

    I also find this example unconvincing:

    “A train is speeding down the hill where it will certainly crash and kill all aboard unless a switch is thrown which will divert the train to another track, where it will certainly kill a man standing on that track. You are at the switch: what do you do?”

    It simply is not very comparable, the situation is far more concrete; and neither conclusion involves deliberate killing.

  31. Bernie

    I think you presume a given value of the death of someone killed by the state is somehow more valuable than that of an innocent murdered by someone not deterred because of the absence of the death penalty. The question would be if I said that there is clear statistical evidence that for every 1 innocent person killed by the state 10 murders would be averted by virtue of the presence of the death penalty, would that be sufficient for you to sanction the death penalty. If not, what number would suffice? The same utilitarian logic can be applied to the train switch example. Of course you could argue that even if 5000 were on the train you still would not sacrifice the one. Alternatively you might pose the counter question that the death penalty leads to more murders and therefore we should eliminate the death penalty and sacrifice whatever immediate deterrent value the death penalty has in order to change the “culture of violence” to a longer term reduction in murders overall. The challenge in both positions is to provide evidence and data that bears on the asserted empirical relationships.

  32. Gary Moran


    I think you are replying to me?

    In the case of the equality of innocents, it isn’t that possible innocents killed by the state are more valuable, rather that the act of execution is more concrete. You can’t justify killing innocent citizens, by balancing them against the greater number of probable innocents that might be killed later. If you could, then you might be able to justify executing sociopaths before they have evolved into serial killers.

    As for the train example, I would likely pull the switch whereby the one would be killed. I simply didn’t think it was analogous enough to the execution question.

  33. Bernie

    You are correct I was responding to you.

    Now I do not understand: How is the possible execution of innocents more “concrete” that the possible murders of those not deterred because of the absence of the death penalty. That is like saying that a probability of 50% is more concrete than a probability of 20%. I would agree that there is more certainty to the execution of innocents than the murder of innocents by the undeterred but surely that is off-set by the fact that it is non-zero and there are large numbers to be deterred.

    Perhaps Matt can restate this in Bayesian terms so we can size the issue.

  34. Gary Moran


    it is more concrete because the decision making is on behalf of the state and it would be the state or its proxy that would be doing the killing.

    When thinking about the acceptability of killing people, its not just the death itself that is weighed, but the action of the killer and the intent. So someone who fails to save a drowning person has committed no offense, someone who kills by reckless driving may only have committed a civil offense and might be fined, while a serial killer would be incarcerated and possibly executed.

    Therefore it is my contention that the state deliberately killing those convicted of criminal offenses should, in the case of innocents unintentionally slayed, be weighed greater than those killed by murderers not deterred because the death penalty was not in place.

    Another way of looking at it is that the state cannot be blamed for the crimes of others; that it is impossible for the state to remove evil from our lives, and that state sanctions to improve our collective lives should not be at the expense of our liberties or our morals. Quite what our liberties and morals are is more subjective – but we should all be able to recognize that consequences by action and inaction are not the same.

  35. Gary Moran


    sorry I wrote all that and realized it doesn’t answer the question of why its more concrete – I’ll try again.

    In the train example given: you are balancing 1 probable death caused by your action intended to save 100 lives, against 100 probable deaths caused by your inaction. But in either scenario its you, you have a good idea of what the possible consequences are, in both cases the deaths occur reasonably quickly, and there is a definite link between your actions and the consequences.

    In the death penalty case its probable innocent deaths committed by you (the state) now, balanced against probable deaths committed by others in the future. Furthermore there is no definite link between your actions and your intended consequences. You are balancing actual murder against possible lives saved. Which is why I say the lives lost by execution are more concrete.

  36. Spence_UK

    Regarding the train track experiment, this is an interesting one; the work I’m aware of was conducted by Marc Hauser, and was actually used by him to justify a biological / evolutionary explanation for morality (although this is not widely accepted as no mechanism is known for it), very much along the lines of the “selfish gene”.

    However, putting the original purpose of the work aside, we’re missing an important point about the train track experiment. There were several different variants of the moral dilemma, teasing out consistent moral views across a wide section of society (age, religion etc.).

    In one example – an unmanned trolley runs along the tracks, and will hit five people. Points can be switched by an individual, to send the trolley into a siding. An innocent bystander will be hit by the trolley in the siding. This is dilemma one; most agreed (>90%) that the points should be switched, killing the one but saving the five.

    In another test, a fat man sits on a bridge, with a trolley heading towards the five people. By pushing the fat man off the bridge, the trolley would be stopped, saving the five people. In this instance, virtually nobody agreed (<10%) that the fat man should be used to save the five people, despite the remarkable similarity in terms of outcomes to the first dilemma.

    It seems people have a built in code that says "accidental collateral" is acceptable, but when we actively use someone as part of the solution (so to speak), it is unacceptable. Another good example is where five people need organ transplants and they will die, it would be immoral to kill one healthy person to provide the five organs. This is once again consistent; even though more would be saved, the direct involvement of the individual seems to hit a built-in moral stop.

    In this respect, it seems to me that executing someone as a future deterrent rather falls into the latter category, i.e. that most people find immoral. Of course, it may just be that biology or evolution has programmed this moral code into us, and the is-ought logical leap applies; just because nature has made us this way, doesn't mean this is how things should be.

    Just because some group have determined that execution is "lawful" killing doesn't make it any less barbaric. If you wish to make an assessment of barbarism in society, executing an individual that poses no direct risk to society is (IMHO) no less barbaric than someone committing murder.

    Also, executions sacrifice personal human rights in favour of state protection. Yuck. Those who sacrifice personal liberty for safety deserve neither liberty nor safety.

  37. Bernie

    Gary & Spence:
    Interesting. I will have to get back to you. As to the the fat man and the organ transplant dilemmas – and whether there is some biological code – I doubt it. Military guys need to make these types of decisions all the time. The honorable ones with regret: The saddest thing after a battle lost, is a battle won. (Wellington at Waterloo)

  38. Gary Moran


    very illuminating, thanks.

Leave a Reply

Your email address will not be published. Required fields are marked *