My review of Alice Goffman’s On The Run: Fugitive Life in an American City was posted today on The New Rambler Review (an online journal edited by Eric Posner, Adrian Vermeule, and Blakey Vermeule). As you will see, I am extremely critical of the book – especially on ethics issues – despite the raves from luminaries in the world of social science.
You can read the entire review here.
I have copied a few key paragraphs below, which will allow you to get an idea of the whole thing:
Alice Goffman’s widely acclaimed On the Run: Fugitive Life in an American City has drawn more positive attention than almost any sociology book in recent years. The success of the book led to a lecture tour of at least twenty sociology departments and conferences. Her TED talk, which was often interrupted by applause, has had nearly 700,000 views. A careful reading of On the Run, however, leaves me with vexing questions about the author’s accuracy and reliability. There are just too many incidents that strike me as unlikely to have occurred as she describes them. One must try to keep an open mind about such things – especially regarding someone as obviously brilliant and dedicated as Goffman – so readers may disagree with me about the extent of her embellishments. In any event, there is a bigger problem. As I will explain below, Goffman appears to have participated in a serious felony in the course of her field work – a circumstance that seems to have escaped the notice of her teachers, her mentors, her publishers, her admirers, and even her critics.
More after the jump.
*
There is another reason for concern about the hospital story. Elsewhere in the book, Goffman explains that many of her subjects refused to seek medical attention, or visit sick or injured friends, for fear that their names would be run by the cops. While it is understandable that the police might check out the emergency room for patients with gunshot wounds, I believe it is an urban legend that they likewise screen all patients and visitors in every ward. I found no one else who ever heard of such a routine practice.
By validating the rumor, however, Goffman has now embedded it in ethnographic lore, and it could well be accepted as fact by sociology and social work majors. If repeated uncritically by future social workers in urban areas, this could have the ripple effect of further discouraging young African-American men from obtaining necessary medical care, which would be a shame.
*
Taking Goffman’s narrative at face value, one would have to conclude that her actions – driving around with an armed man, looking for somebody to kill – constituted conspiracy to commit murder under Pennsylvania law. In the language of the applicable statute, she agreed to aid another person “in the planning or commission” of a crime – in this case, murder. As with other “inchoate” crimes, the offense of conspiracy is completed simply by the agreement itself and the subsequent commission of a single “overt act” in furtherance of the crime, such as voluntarily driving the getaway car.
I sent the relevant paragraphs from On the Run to four current or former prosecutors with experience in Pennsylvania, New York, New Jersey, and Illinois. Their unanimous opinion was that Goffman had committed a felony. A former prosecutor from the Philadelphia District Attorney’s office was typical of the group. “She's flat out confessed to conspiring to commit murder and could be charged and convicted based on this account right now,” he said.
*
Medical students are taught to do no harm. Law students are instructed that they may not assist a client in the commission of a crime. The analog for ethnography students ought to be equally straightforward: if a subject asks you for help in a murder plot, just say no.
This review is part of a longer project on the ethics of ethnography.
"While it is understandable that the police might check out the emergency room for patients with gunshot wounds, I believe it is an urban legend that they likewise screen all patients and visitors in every ward. I found no one else who ever heard of such a routine practice."
Does it really matter from an ethnographic standpoint? Behavior is determined by belief, not necessarily objective fact. Ethnography is mainly getting inside a culture and trying to see what they see.
Posted by: twbb | May 26, 2015 at 02:52 PM
You are right, of course, that it would not matter if Goffman had reported it as a belief -- which indeed it is. But she reported it as a fact.
That's sort of like doing ethnography among Tea Partiers in Texas, and reporting, on that basis, that Operation Jade Helm is actually hiding weapons in tunnels beneath abandoned Walmart locations.
Posted by: SL | May 26, 2015 at 03:00 PM
I'm puzzled by this post. It is exceptional for a US state to lack a requirement for mandatory reporting of gun-shot wounds. Its actually faster to list the ones that don't - Alabama, Connecticut, Indiana, Kentucky (but not totally), Louisiana, New Mexico, Oklahoma (except for criminal injury), PA, South Carolina, South Dakota, Washington and Wyoming. All other states and Washington DC mandate reporting of gunshot wounds.
PA is one of the few where doctors do not have to report.
Posted by: [M][@][c][K] | May 26, 2015 at 03:19 PM
That is fascinating, brackets. Do you have a source? It makes Goffman's story even less likely.
Posted by: SL | May 26, 2015 at 03:37 PM
I just ran a quick search - website on domestic violence law had a state by state listing, with the state statutes cited.
Posted by: [M][@][c][K] | May 26, 2015 at 03:45 PM
This is fascinating from a research ethics perspective in many ways, Steve; thanks for writing and sharing it. I haven't read the book. Does Goffman say that her IRB required her to destroy her notes? It would not surprise me if her IRB did require her to destroy her notes containing details of her subjects' criminal activity and frankly there are significant risks to subjects if those notes got into the wrong hands (although, as you suggest, there are far less drastic ways to protect subjects than wholesale destruction of all notes). On the other hand, given how intertwined she became with her subjects by the end, it also would not surprise me to learn that she voluntarily chose to destroy her notes to protect them.
(Compare this to the LaCour same-sex marriage canvassing study situation. The NYT reports that he told the editor of Science through his lawyer that he destroyed the data he is now alleged to have fabricated in order to protect subjects. I seriously doubt either that his IRB required him to destroy fairly innocuous survey data or that he took it upon himself to destroy his data to protect subjects. Also, prior reports have him telling his co-author that he deleted it by accident.)
Posted by: Michelle Meyer | May 26, 2015 at 10:05 PM
"I spoke with a . . . current Philadelphia prosecutor"
"I sent the relevant paragraphs of On the Run to a source in the Philadelphia Police Department "
"I sent the relevant paragraphs from On the Run to four current or former prosecutors with experience in Pennsylvania, New York, New Jersey, and Illinois. Their unanimous opinion was that Goffman had committed a felony. A former prosecutor from the Philadelphia District Attorney’s office was typical of the group. “She's flat out confessed to conspiring to commit murder and could be charged and convicted based on this account right now,” he said."
Geeze, Steve, you might not be trying to "censure" her, but you certainly seem trying to get her arrested...
I'm kidding, I'm kidding; her fieldwork was done long enough ago that I think it's fairly uncontroversial that any charges would be barred by the statute of limitations.
Posted by: twbb | May 26, 2015 at 10:22 PM
twbb,
Don't kid. If any of her alleged "conspiracy to commit murder" constituting actions were coupled with someone actually getting killed, then there is no statute of limitations for it in Penn. 42 Pa. Code Sec. 5551. If no one got killed, the statute would be five years plus tolling from any period she lived out of state or during which her alleged co-conspirators we prosecuted. 42 Pa. Code Sec. 5552, 5554. Having not read the book and being unfamiliar with the details, I have no idea if she's outside that window.
Posted by: Former Editor | May 27, 2015 at 09:34 AM
Hmmm, good point though I was assuming that if a murder actually happened it would have been mentioned at least in Steve's review; I was assuming the 5-year limitations period was applying here. I was not aware, however, of the out-of-state tolling issue though (I have absolutely no background in Pennsylvania law). In that case the review does look a little like it's kind of nudging prosecutors to do something...
Posted by: twbb | May 27, 2015 at 10:25 AM
This is outside my area of expertise too, but I thought it was worth mentioning. From the way the article describes it, her driving the car was to assist in a coordinated effort to engage in a revenge killing against a specific gang member. It seems more than possible to me that the author's alleged activities in connection with that effort would implicate her in a broader murder conspiracy, even if the actual murder of that gang member did not take place on an evening when she was directly helping and which she does not describe in the book.
Posted by: Former Editor | May 27, 2015 at 11:18 AM
Probably would be a hard case to make; the only evidence they'd have would be the book unless they could somehow track down the relevant actors, and she could pretty easily construct an interpretation of the events in the book that took her outside the criminal conspiracy statute, including just saying she made it up.
Posted by: twbb | May 27, 2015 at 12:15 PM
twbb - as far as I know there is no "statute of limitations" on murder - it is certainly very unusual. Some states have a limit though on conspiracy to commit murder.
Posted by: [M][@][c][K] | May 27, 2015 at 02:55 PM
There was no murder. She tells the story only of a conspiracy.
Posted by: SL | May 27, 2015 at 03:00 PM
As I read the code, the SOL in Pennsylvania for conspiracy to commit murder is five years if no murder occurs. Although the odds that a prosecution would be brought in these circumstances seem vanishingly small even if we were within the SOL.
Posted by: Orin Kerr | May 27, 2015 at 03:15 PM
I agree with Orin and twbb that prosecution is extremely unlikely even if there is an unexpired SOL or no SOL. I was only pointing out that it's not a foregone conclusion that the author is protected by the SOL should one of the prosecutors Steve mentions contacting decide the issue is worth looking into.
Posted by: Former Editor | May 27, 2015 at 04:00 PM
It's doubtful if the story of the car ride is even true. In many ways, Alice Goffman is the new Margaret Mead
Posted by: Enrique Guerra-Pujol | May 27, 2015 at 11:13 PM
[Mack]: It's one thing to have mandatory reporting of gunshot wounds, and another thing for hospitals to give the names of visitors in the maternity ward to the police, which is what Lubet quotes Goffman as writing about.
Posted by: Joshua | May 31, 2015 at 01:24 PM