As I explained in this earlier post, I recently had an exchange with Berkeley's Michael Burawoy, published in Contexts, on the question of accuracy in ethnography. My initial essay is here; Burawoy's disapproving response is here. I have now posted Part Three in the exchange, my reply to Burawoy, which I have copied in full below:
Ethnographers in Cars with Guns
On the last day of her life, December 30, 2018, seven-year-old Jazmine Barnes was accompanying her mother and two younger sisters on an early morning run to a Houston convenience store. Suddenly, another automobile pulled alongside their car and someone began shooting at them. “As I turned around and looked back at the street, I heard shots start firing and they came through my window, broke my glass, and hit me in my arm,” said Jazmine’s mother, LaPorsha Washington. The shots kept coming as the other car sped away. “Momma, Jazmine’s not moving,” cried one of the girls. ”She’s not talking.” Washington turned around and realized, “my 7-year-old was shot in the head.” Jazmine died at the scene.
Two men have been arrested and charged in what was evidently a case of mistaken identity, a revenge shooting. It should be no surprise that both the shooter and his driver have been indicted for murder. Under Texas law, as in every state, anyone who participates in a crime can be prosecuted for the consequences.
The victims of drive-by shootings are not always in other cars. In January 2013, fifteen-year-old Hadiya Pendleton proudly marched with her high school band in President Obama’s second inauguration. One week later, she was standing with classmates in a Chicago park, taking a break from school following an honors class exam. The students were somehow mistaken for gang members, and Hadiya was killed in an all too common rivalry shooting. Two young men have been convicted Hadiya’s murder – the one who fired the gun, and another who helped him plan it.
These stories could be multiplied by thousands over recent years, with bystander victims of all ages gunned down in parks, alleys, stores, schoolyards, and even their own homes. It takes no great insight to recognize that everyone responsible for these crimes – both the shooters and their enablers – should be held morally and legally accountable.
This brings us, I regret to say, to the case of Alice Goffman, as raised by Michael Burawoy. I had no intention of revisiting this issue, and I trust that Contexts readers have noticed that Goffman’s name does not appear in my opening essay (in the Winter 2019 issue). Unfortunately, Burawoy insists that Goffman’s participation in a would-be drive-by shooting, as described in her Methodological Appendix to On the Run, was all in a day’s ethnographic work. This requires a somewhat detailed response, so please bear with me.
Here is how Goffman recounts the manhunt that followed the murder of her friend, Chuck:
Many nights, Mike and Steve drove around looking for the shooter. . . . On a few of these nights, Mike had nobody to ride along with him, so I volunteered. We started out around 3:00 a.m., with Mike in the passenger seat, his hand on his Glock as he directed me around the area. We peered into dark houses and looked at license plates and car models as Mike spoke on the phone with others who had information about the 4th Street Boys’ whereabouts.
One night Mike thought he saw a 4th Street guy walk into a Chinese restaurant. He tucked his gun in his jeans, got out of the car, and hid in the adjacent alleyway. I waited in the car with the engine running read to speed off as soon as Mike ran back and got inside.
Luckily, Mike realized that he had the wrong man, and nobody was shot that night.
As I have noted before, these activities constituted conspiracy to commit murder under the law of every U.S. jurisdiction.
Burawoy, however, sees nothing wrong with an ethnographer’s participation in what he himself characterizes as a “murderous pursuit”:
If you are studying people who commit crimes, you are likely to get involved in those crimes. To maintain relations with the people you study, you often have to do what they do.
This rationalization is untenable. There is nothing about ethnographic research that warrants or excuses, much less requires, actively endangering the lives and safety of others. Randol Contreras and Phillipe Bourgois managed to “maintain relations” with subjects who were torturers and rapists, without getting “involved in those crimes.” I doubt that Burawoy would adhere to his cavalier position for ethnographers who studied, say, racists who were plotting to murder Somali immigrants. (Three Kansas men were recently convicted of such a conspiracy, even though the planned bombing was never carried out.)
Burawoy bolsters his argument by informing us that he has also “transgressed the law” by participating in an anti-Apartheid demonstration while conducting research in Zambia. The comparison to a drive-by shooting is offensive. There is a vast difference between protesting South African racial policies and jeopardizing innocent lives. The acceptability of the former, which is morally blameless although locally illegal, does not begin to justify the latter.
This distinction is well understood in jurisprudence and moral philosophy, and I would have hoped in sociology. It is the difference between an act that is malum prohibitum, which is wrong only because it is prohibited, and one that is malum in se, which is inherently bad, in and of itself. As I explained in Interrogating Ethnography, it is reasonable for ethnographers to commit victimless crimes (as when Howard Becker smoked weed with jazz musicians) or to violate cruel and discriminatory laws (as in Laud Humphreys’ research on homosexuality). Alice Goffman’s potentially lethal ambush falls into quite another category.
In the summer of 2007, when the events in question occurred, 108 African Americans were murdered in Philadelphia, mostly young men, 94 of them by gunfire. Included were two murders that had already occurred in what Goffman called the “Fourth Street War.” Her “Glock ride,” though ultimately unsuccessful, threatened to provoke another round in the cycle of deadly retribution. She was not a ride-along observer, but the volunteer driver when no one else was available – meaning that the manhunt would not have happened on those nights without her. Ethnography can be socially useful, but it is not comparable to, say, medical research, where calculated risks can lead to great benefits. Whatever insights Goffman might have gleaned from her ambush attempt, they were not nearly worth endangering bystanders in crossfire. Even her own dissertation advisor, Princeton’s Mitchell Duneier, recognized that she “crossed an ethical line.”
Few would fail to appreciate the difference between joining a political demonstration and plotting a potential revenge killing. But in case there is doubt, imagine what would happen if a graduate student were to consult her advisor, or an IRB, before taking the wheel. Would anyone ever give advance approval to an armed manhunt on the theory that it would help “maintain relations” with a research subject? And if so, how could that be explained to the heartbroken parents of Jazmine Barnes and Hadiya Pendleton?
Burawoy has other complaints about my essay and book, most of which mischaracterize my writing almost beyond recognition. He begins by asserting that I engage in “ferreting out random errors in monographs to discredit them” and thus, “if any of the facts are false, ipso facto, the entire theory is false” and its “contribution is zero (or even negative).” (Italics original.) I have said nothing of the sort. My purpose, as I repeatedly state in both essay and book, is to demonstrate ways in which ethnography can be strengthened through enhanced accuracy. I praise many of the monographs that I have fact checked, including aspects of On the Run. Burawoy seems to think I intended to “discredit” Edin and Shaefer’s $2 a Day, which I actually describe as “powerful and compelling” and “a thorough and closely documented account of extreme poverty,” while explaining that my critique is meant to illustrate the uses of circumstantial evidence, rather than to challenge the overall argument of the book.
Elsewhere, Burawoy achieves the same distorted result through elision. Quoting my observation that documentary evidence is “frozen in time, unlike fragile human memories that may change with every retelling,” he criticizes me because “being frozen in time doesn’t make a piece of evidence more reliable.” He conveniently omits my cautions about documentary evidence, where I set out the circumstances in which it can be either more or less trustworthy. I state specifically that documents are not always accurate and unbiased, explaining only that they have “certain advantages over human memory, and should therefore be consulted if available.” One hopes that every careful researcher would agree, but Burawoy seems determined to recast my views as tendentiously as possible.
In that mode, Burawoy continues to accept Goffman’s claim that the Philadelphia police “wait outside hospitals serving poor Black communities and run the IDs of the men walking inside,” while also examining hospital records for suspects. This is an alleged phenomenon that Goffman alone claims to have observed. No other person – including ethnographers working in the same communities and investigative journalists – has ever reported such incidents at Philadelphia hospitals. Burawoy complains that I give credence to “hospital administrators, former public defenders, and police,” but how else would one attempt to confirm or disconfirm Goffman’s unique assertion? To Burawoy, the unsourced statement of a single ethnographer – claiming, inter alia, to have witnessed three maternity floor arrests in one evening – is sufficient to create a “field of contestation” that can never be resolved. Thus, no ethnographer could ever be shown wrong, no matter how much evidence is produced to the contrary.
Burawoy notes that “an abiding motivation for ethnographic research is to contest ‘official’ views of the world,” which explains his result-oriented deference to an unsupported account about unverifiable events, which just happens to comport with his world-view. In the words of the psychologist Thomas Gilovich,
For propositions we want to believe, we ask only that the evidence not force us to believe otherwise. . . . For propositions we want to resist, however, we ask whether the evidence compels such a distasteful conclusion. (Gilovich, How We Know What Isn’t So, pp. 83-84, italics original.)
In contrast, my own research was open-ended. I would have reported finding any corroboration for Goffman’s story, as I have done in other cases, recounted in Interrogating Ethnography, where my initial misgivings turned out to be in error.
In any event, Burawoy ignores the fact that no sociologist questioned the hospital story – which he now concedes is contestable – until I did the countervailing research. Nonetheless, he accuses me of acting in bad faith (and worse), by joining what he calls a “social movement to pillory” Goffman. “It is probably no accident,” he opines, that I have “target[ed] a young scholar.” Although there is no such social movement, much less a pillory, it was indeed no accident that my attention was drawn to On the Run. And why not? It was an acclaimed best-seller, the subject of an auction for the paperback rights, the recipient of numerous glowing reviews. It was included on the New York Times list of the most important books of the year, and endorsed and promoted by many senior figures in sociology. Are we allowed only to praise such books? Or do scholarly standards call for the same scrutiny as every other prominent and widely publicized work?
If there is an explanation for so much ax-grinding, it is most likely that I have unintentionally stepped into the middle of a profound and seemingly bitter debate between empiricists and “theory-driven” ethnographers, with Burawoy aggressively in the latter camp. As an outsider to this disputation, I would have thought that accuracy and reliability would be equally important to both sides, but Burawoy sees it differently. He associates me with the dread empiricists, who have “hijacked” ethnography in pursuit of “a strategy of power to subjugate young insurgents within a contested field.”
Hence, the risible accusation that I have become “like Mike, carrying a metaphorical gun, hunting for the killer-error in every nook.” To Burawoy, I guess, real guns are just fine for ethnographers but fact-checking can be deadly.
I know it's kind of tangential to your response to Burawoy, but I'm curious about your assertion that Goffman conspired with Mike to commit murder. Conspiracy requires a double intent: the intent to agree and the intent to commit the object of the conspiracy. I don't see how Goffman had either intent in the scenario above — particularly not the intent for Mike to actually kill the shooter. To be sure, had Mike actually killed someone, Goffman could easily have been convicted of aiding and abetting that murder. But I don't see the argument for conspiracy as an inchoate crime.
First, let me repeat that I had no intention of returning to Goffman's murder conspiracy, and my initial Contexts essay did not mention her at all. I addressed it in Part III only because Burawoy defended the "Glock ride" as acceptable, and perhaps even admirable, ethnography.
I suppose that now obliges me to answer Kevin Heller's very reasonable question.
In Pennsylvania, and most U.S. states, a conspiracy is complete when there has been agreement to commit the crime plus an overt act. Goffman wrote that she agreed the drive the car "because I wanted Chuck's killer to die." So there you have agreement. Driving the getaway car — especially when Mike followed a suspected victim into an alley — was an overt act.
In Interrogating Ethnography, I provided the full vignette to four former prosecutors in state and federal courts, including Pennsylvania, and they agreed unanimously that Goffman had committed conspiracy to commit murder.
For an example of a recent federal prosecution (and conviction) based on agreement plus overt act, but in which the crime itself was never carried out, see: https://www.justice.gov/opa/pr/three-southwest-kansas-men-sentenced-prison-plotting-bomb-somali-immigrants-garden-city
I want to share some thoughts provoked by this: “There is nothing about ethnographic research that warrants or excuses, much less requires, actively endangering the lives and safety of others. Randol Contreras and Phillipe Bourgois managed to ‘maintain relations’ with subjects who were torturers and rapists, without getting ‘involved in those crimes.’ I doubt that Burawoy would adhere to his cavalier position for ethnographers who studied, say, racists who were plotting to murder Somali immigrants.”
While I agree with Steve, I would like to provide an example in which individuals (as members of groups) do and probably must maintain relations with those who have and are continuing to commit morally reprehensible acts and crimes (and Steve may very well be aware of these other and cases that are significantly different to that in the instant case) and in which the relevant moral considerations and quandaries (some would say paradoxes) are not captured by the otherwise important distinction “between an act that is malum prohibitum, which is wrong only because it is prohibited, and one that is malum in se, which is inherently bad, in and of itself,” in other words, this distinction is not helpful in or at least central to addressing the morally relevant or salient features or properties of actions and their consequences in these cases. The cases I am referring to involve individuals, in their capacity as group members, are deliberately maintaining or even cultivating relations with morally bad or evil actors. More importantly, such cases find sufficient moral warrant or excuse or morally mitigating factors that help us see why the maintenance of such relations is, however regrettably, necessary, if only because of the moral good that results alongside of or in spite of the moral bad or evil that is not directly confronted or eliminated.
This exemplum thus does not involve ethnographic research (or social scientific research of any sort) but humanitarian aid workers in situations of violent conflict in which the overarching aim is to prevent and or/eliminate the suffering of civilians that has arisen in whole or in part* owing to that conflict; perhaps more precisely and generally or ideally speaking: “The aim of humanitarian action is its own disappearance. Or rather, it is the disappearance of the conditions occasioning the need for humanitarian assistance in the first place.” The example I draw upon is from a very important book: Chiara Lepora and Robert E. Goodin’s On Complicity and Compromise (Oxford University Press, 2013).
The wrongs in this instance are morally (i.e., inherently) bad, and the humanitarian aid workers are in some respects and minimally speaking, clearly complicit (which is rightly distinguished from it ‘conceptual cousins,’ namely, connivance, contiguity, collusion, collaboration, condoning, consorting, conspiring, and full joint wrongdoing’) in the ongoing commission of at once illegal and immoral acts (the former minimally viewed from the vantage point of international humanitarian law and human rights), thus they are in some undeniable and troubling ways “involved in [these] crimes.” In this instance, “genuinely well-meaning people, with great reluctance and regret, find themselves contributing to the wrongdoing of others because that is the only way that they themselves can accomplish some greater good.” There is here a question of moral judgment and a weighing of consequences in which one is knowingly but regretfully complicit in acts otherwise deemed malum in se, but only because such complicity allows for acts of compassion and beneficence that would not otherwise occur in absence of this complicity, and the resulting good is considered sufficient to outweigh the otherwise morally bad action. This is morally hazardous territory, to be sure, for moral acts in the real world are often messy and thus with some justification we have a body of literature on the occasional or “obligatory” need for—in even conventional political environments—“dirty hands” and yet, as C.A.J. (Tony) has argued, not all such putative cases are correctly characterized as necessary or unavoidable, particularly when the conditions that give rise to a perception of unavoidability or necessity are mutable, as in those cases, often commonplace, when political environments are morally bankrupt or corrupt. Coady contends that not only are we
“likely to draw the wrong norms from politically behavior [in effect, inferring from what we conveniently believe to be—contingently or relatively—the case, situation or circumstance, to what normatively should be the case with regard to our political behavior] but that we tend to focus our moral concerns too narrowly. We concentrate upon the particular act that will require dirty hands and ignore the contingency and mutability of the circumstance that have given rise to it. Yet it is precisely these circumstances which often most deserve moral scrutiny and criticism, and the changes which may result from such criticism [which is not to that claim criticism alone will accomplish such changes, only that it sets in motion, if you will, the causal chain that leads to the requisite changes] can eliminate the ‘necessity’ for those types of dirty hands in the future. [….] Robert Fullinwider once remarked that we need politicians just as we need garbage collectors, and in both cases we should expect them to stink. But, once upon a time, we needed the collectors of what was euphemistically called ‘night soil’ and, in many parts of the world, human ingenuity has eliminated the need for that very malodorous occupation [a job which, at least in some parts of the world, is still thought to be fit for people who are themselves said to be categorically “polluted,” a religiously and socially sanctioned inhumane form of status, standing, and treatment that denies such individuals in these groups full personhood and thus exercise of the powers and capacities we understand as fundamental to human agency as well as, and relatedly, essential to the recognition of and respect for human dignity].”
But in the case of these humanitarian aid workers (in situations in some respects similar to those undercover operations in law enforcement where, in order to gain the full trust of the individuals involved, say, in gun trafficking or the trafficking of human beings, may act in varying degrees of complicity or collaboration with those they are seeking to eventually arrest and convict for their crimes) there is no immediate or realistic imminent prospect for altering let alone removing the contingent circumstances or conditions and thus their only real hope for aiding in the relief of suffering may entail moral entanglement or cooperation with those committing morally egregious if not evil acts (e.g., genocide or other ‘crimes against humanity’). Because it would take up too much time and space and tax most readers’ attention span, I will not illustrate these examples but refer the interested reader to Lepora and Goodin’s compelling discussion and analysis of the situation in the refugee camps that were established in the Great Lakes region of Africa (1994-1996) as a result of the exodus of refugees, largely Hutu, who were passively or actively opposed to the genocide against the Rwandan Tutsi minority: “Having first committed genocide against the Tutsi, the FAR (Rwandan Armed Forces) soon began using the refugee (mainly Hutu) population in various ways: as a source of income and power in their own right; as a lure for international assistance and legitimation; and as a protection against those who might punish or retaliate against the FAR for the genocide.” Put more starkly, “Those refugee camps [largely established or organized by FAR!] constituted … a terrible and consistent example of a human shield for the military and political perpetrators of genocide. The use of human shields is, of course, prohibited explicitly as a war crime under international law.” Among the conspicuous horrors associated with the camps was a cholera epidemic in 1994 that caused “more than 80,000 deaths in ten days” and continued to ravage the camps for months.
As for the moral quandary and related dilemmas that arose from these refugee camps the following will have to suffice, for I cannot do justice to the incisive and fairly comprehensive nature of the analysis and real world exemplum provided by Lepora and Goodin:
“When providing aid, [international and regional humanitarian] organizations [NGOs] were obliged to acknowledge, interact with and contribute to those perpetrators of genocide. [The NGOs] that intervened on a purely humanitarian basis thus ended up contributing to FAR’s power, from a symbolic and sometimes material point of view. All international aid organizations faced the same dilemma: continue working in the camp, and thereby strengthen further the power of genocidal perpetrators over the refugees; or withdraw from the camps, abandoning a population that was in extreme distress.”
* I suspect it is not always necessary or even possible to draw a distinct line between the suffering that civilians were experiencing prior to a violent conflict and that suffering said to be directly or indirectly caused by that conflict, even if it is clear that conditions characterized by needless or unwanted and eliminable suffering have worsened in the latter case.