After the dust has settled and the smoke has cleared a bit from the 521 incident on the Taipei MRT that left four people dead, over twenty people injured, and a psychopath in custody, I thought it might be time for me to weigh in on the tragedy. I will do so in response to an article which triggered a rethinking of my own position on capital punishment.
On May 25, Thinking Taiwan published an article titled “What We Can Learn From Mass Murderers,” which can be accessed here. Although thought-provoking due to its departure from the moral diatribes that tend to dominate the discussions on capital punishment, and thus a welcomed fresh perspective, “What We Can Learn” is also problematic.
The central argument of the piece, briefly, is that Cheng Chieh (the accused perpetrator) should not be put to death in an “emotion[al] . . . desire for vengeance” but should be kept alive in order to become a lab rat. On the surface, this argument seems both logical and practical; the article itself uses the example of Hannibal Lecter from Thomas Harris’ Silence of the Lambs to drive the point home. But three principal problems appear in the article.
First, the article runs through a list of psychological and behavioral traits common to both mass murderers and serial killers: “sociopathic behavior, an accompanying sense of being victimized by society, a strong narcissistic streak, and an inability to empathize,” “[t]heir [mass murderers and serial killers] impulses were driven by an ‘us’ versus ‘them’ mentality and a narcissistic desire for recognition, to leave their mark in history,” and the feeling among “[m]ost mass killers in the non-terrorist/genocidal/familicide category [that] they needed to do ‘something big’ . . . .” It also runs through several common causal factors for such behaviors: “An abusive family environment, a nagging/promiscuous mother, or bullying in the schoolyard, among others, are recurrent red flags.” It also lists several subcategories of mass murder: “terrorism (killing for political or nihilistic purposes, e.g., al-Qaeda and Aum Shinrikyo), familicide (attainment of power, e.g., the Nepalese royal family massacre), or genocide (the extermination of a race, again for predominantly political purposes, e.g., Rwanda).”
It seems a great deal is already known about the average mass murderer: one can even list subcategories of mass murder, causal factors, and personality and behavioral traits. This demands the question, however: if so much is already known about mass murderers (and serial killers), why must one particular mass murderer (as the case may be) be kept alive to be studied? One need not study all individual apples to understand the apple.
But perhaps we are comparing apples to oranges? This possibility may be a way out of the conundrum presented above, and it may salvage the article’s main argument. Here again, however, we run into problems. First, if we focus on individual mass murderers, then the above subcategories, causal factors, and personality and behavioral traits are rendered meaningless because they lack relationships between individual mass murderers. More formally, they are not common independent variables shared by a group of dependent variables; they are simply random data. This may be interesting for the biographer, but in attempting to form a theory of social behavior and psychology, they are utterly useless until some relationship can be found.
Second, if these factors (or variables) are indeed random and particular to the individual and the context of the act, and if there is no overarching relationship between shared independent and dependent variables (i.e., if there is no pattern), then keeping Cheng alive teaches us nothing unless we seek to write a biography or seek to, for whatever reason, become interested in his life. In this sense, study becomes a more professional form of fetish. One should study all species of fruit simply because one enjoys studying fruit species; there are no two apples. Study becomes “intellectual curiosity” that the article argues it need not become, or study becomes potentially an art. Either way, very little, if any, benefit can be derived from sparing Cheng.
Third, it is possible to argue that one might study Cheng because he undoubtedly fits into both categories: likely, his personality fits somewhere into both the general and the specific categories of study, listed above. This is usually the realm of inquiry that all scholarly inquiry tends to fall under. Cheng is an apple, essentially, but there are certain variations to this particular apple that must be explained, as they do not seem to fit precisely into any previously constructed theory of apples. This is the article’s best argument, but inexplicably the article then rules this possibility out by stating that such criminals are a) “not [capable of] weigh[ing] costs and benefits along rational lines” and b) useful because “investigators (like FBI agent Clarice Starling in Silence of the Lambs) often turn to psychopaths to help identify serial killers on the loose, because they, above all, are cognizant of the things that make them tick. In the absence of such sources, we can fall back on scientific models, which can only be established if we study killers like Cheng.”
The problem here is twofold. First, it is not that mass murderers are not rational; I think it is safe to say that cognitive psychology has ruled out the possibility of any one form of rationality. Moreover, even “rationality” itself has variants: what cost I weigh against what benefit and what weight I give to particular costs and benefits are extremely subjective. What person A interprets as a costly gamble person B might consider a viable option, person C a worst-case scenario, and person D a wonderful opportunity—and that’s just in a four-person game. Say what one will about cognitive psychology; it does not, of course, completely destroy the idea of the rational actor. But rationality, no matter how it is defined, cannot explain why “rational” people behave differently in the same situation. Some intervening variable has to be inserted—for example, a lack of information, time constraints, lack of sleep, etc.—to operationalize the rational-actor model. It is not, therefore, that mass murderers are not rational per se; it is that variations in rationality exist.
Second, and perhaps most problematic of all, if scientific models are something one can merely “fall back on” when resources (i.e., useful mass murderers and serial killers) are scarce, then one has to wonder why all the discussion of personality and behavioral traits common to serial killers and mass murderers needs mentioning at all. Creating scientific models connecting variables to behaviors is how those traits were recognized in the first place. Essentially, then, what the article argues for is keeping serial killers and mass murders alive because they are resources—experts—to use in times of need. But here is the problem: if we reject (or merely “fall back on”) scientific models, then there is no reason to believe that a past serial killer or mass murderer can tell us anything about any current serial killer or mass murderer. Variables are not merely impossible to isolate here; variables cannot exist. It becomes a serious leap of faith: without and theoretical explanation for why one should use serial killers and mass murderers to understand others of their kind, we simply have to believe that they communicate with the same mother ship or supreme sadistic being.
What we come to, then, is three possibilities: first, all sociopaths (there’s a loaded phrase for you) and psychopaths (and there’s another one) are formed from the same mould. No reason to study individual socio- or psychopaths exists. Apples are apples. Period. It is possible to predict and stop all acts of violence.
Second, there exist no such things as socio- or psychopaths. Such terms are merely heuristics. There exist only individuals who behave in disparate ways, and some behaviors are violent. No model can be formed because no relationships between variables exist (indeed, no “variables” exist). Apples are not oranges are not bananas are not kiwi fruits, etc. Period. No acts of violence can be predicted or stopped.
Third, scientific models are imperfect but necessary in locating variations in behaviors across individuals who perpetrate what the layman terms “random acts of violence” (although mass murderers do not consider them “random,” remember). Apples have variations which must be studied and accounted for theoretically if any acts of violence are to be predicted or stopped.
The third possibility is, thus, the strongest, but by rendering scientific models to the background by making them contingent upon individual cases, one undermines the third possibility’s viability. In the end, we are left believing that all serial killers and mass murderers somehow communicate telepathically, yet we lack a rational (read theoretical) explanation for why this may be the case. Interestingly, here we the observers become irrational actors.
This should not be read as a criticism of the article per se. It should be read as an inquiry into why individual criminals merit study. It should also be read as a critique of those who argue that theory construction and model formation are merely exercises in abstraction. I, an opponent of capital punishment under most circumstances, do not agree that Cheng should be kept alive purely for scholarly reasons. Learning things from him may be an unintended advantage of keeping him alive for moral reasons, but it does not follow directly that keeping him alive will produce any scholarly—or social—benefit.
In the end, the greatest hope is simply that such acts never happen again, yet we know as a mathematical fact that they almost invariably will. Thinking Taiwan has shed a different form of light upon this problem—that is, keeping mass murderers and serial killers alive with the hope that they may help us to understand the minds of criminals in order to prevent similar acts from occurring again. It is a fresh perspective that, to be fair to the article’s author (and given my initial thoughts above), probably needs more critical thought than could have been afforded in an initial foray on a new media website. Due to these considerations, the critiques offered here should not be read as outright rejection. The article forced me to revisit deep ontological and epistemological questions that demand addressing in all forms of social-scientific thought. It is my guess that others have approached it in a similar light, and, hence, my main purpose for writing this initial critique.
My condolences go out to the families and loved ones of the slain as well as to those injured and their loved ones. It is my hope that you will be the remembered ones—you and you alone.
 Emphasis in original.
 Emphasis mine. For the circular logic of this statement, see below.
 See note 4.
 Hence the hedged “which can only be established if we study killers like Cheng,” referring to establishing “scientific models,” above. But again there is a problem: it is unlikely that scientific models will ever be foolproof no matter how many individuals are studied (and a theory that attempts to account for all variables is non-falsifiable). There are an infinite number of variables that can influence a model, many of which are discovered after a theoretical model is formed. Hence, we are again not relying on models in this sense but on variations (i.e., individuals). If a model is forever contingent upon individual subjects, then no useful model can be formulated. Possibility number three becomes possibility number two de facto.
Moreover, it is revealing that the article seems to believe that such acts cannot be deterred. One of the criticisms I have heard countless times over the past several days is how Cheng has become something of a “negative celebrity”: we seem to know so much about him now because he did what he did. Interestingly, his stated goal was to commit an act that would be remembered and make him remembered. Perhaps no deterrent can be found herein, but he certainly had a goal and the means to reach it—and he appears to have achieved it. Remembering that much of deterrence is psychological, one might be able to create a theoretical construct in which the ends (fame) are unachievable given the means ([mass] violence). I leave that possibility to social psychological theorists to ponder.
It should also be noted that any action that cannot be deterred also likely cannot be stopped unless preempted. This is a major problem for any free society.
 Or spiteful reasons. That he allegedly has sought the death penalty leads me to believe that if he is not given death, he may one day question his actions and how they so utterly fell short of their goals. See note 4.