Ever been tempted just to cite something that someone else has cited without checking the reference? You know, to say “Wheelbarrow cites Schnittwinkel to show that X is true” without bothering to read Schnittwinkel yourself? Especially if the Schnittwinkel article is in German (or some other language that you don’t read all that well) and you know that it will take you quite some time to make sure that you’ve understood it correctly? Well, I will be tempted no more! Not after yesterday’s research.
There is a very famous paper by Allport and Postman (1945) often cited in the psychological literature about eyewitness testimony that shows how racial stereotyping can influence eyewitness testimony. Allport and Postman showed their participants a picture of an African American man in a suit talking to an Anglo man in overalls, holding a cutthroat razor by his side. They are standing on a subway station. When questioned about it (a week or so later), over half of the participants remembered the razor in the hand of the African American and some of them had him threatening the Anglo with it. Lots of people have cited it, very much as I have.
Except that this is not actually what Allport and Postman did. They actually gave some people a picture showing the event I’ve described above and asked them to describe it while looking at the picture themselves to someone else who couldn’t see it . The second person (still without seeing the picture) then described it to another person, who described it to another, and so on. There were about six people in each “rumour chain” and in over half the rumour chains the razor switched from person to person at some stage in the retellings.
It’s not about eyewitness testimony at all. It’s about the way rumours spread. It was even published first in a paper called The basic psychology of rumor and then in a book called The Psychology of Rumor!
In 1989 (so a mere 44 years after the original) Molly Treadway and Michael McCloskey from Johns Hopkins published a really interesting paper in Applied Cognitive Psychology pointing out that quite a number of people over the intervening four and a bit decades had clearly not read Allport and Postman’s paper very carefully, if at all, because they’d all described the eyewitness version rather than the rumour version. The previous year Julian Boon and Graham Davies from the University of Aberdeen published a paper in the Canadian Journal of Behavioural Science which pointed out the problem without listing those who had mis-cited the original research.
Oh, ooops!! I’m not all that impressed with some of Treadway and McCloskey’s analysis of their attempts to replicate the research that people said Allport and Postman carried out, but I bet there were some red faces when it first hit the streets (or whatever psych journals actually hit).
Fortunately, my regular visits to Mark’s Theological German/Theologisches Deutsch are improving my German reading skills and I can sometimes even skim read German text with understanding, which is also helping to reduce the temptation to report citings unseen, even in German.
And for those who are not good at recognising humour in written form and who haven’t seen my collection of painstakingly gathered journal articles and photocopied book sections, no, I do not make a habit of citing material that I haven’t read. There are just days when I wish I could!