Navigation – Plan du site

AccueilLireLes comptes rendus2017Howard Becker, Evidence

Howard Becker, Evidence

Igor Martinache
Evidence
Howard S. Becker, Evidence, Chicago, University of Chicago Press, 2017, 223 p., ISBN : 9780226466378.
Haut de page

Texte intégral

  • 1 It would be more accurate to say « publishers » insofar as many of them are used to asking research (...)
  • 2 Howard S. Becker, Outsiders: studies in the sociology of deviance, New York, The Free Press of Glen (...)
  • 3 Les ficelles du métier, La Découverte, Paris, 2002.
  • 4 Although his work as a whole contains strong methodological considerations, H. Becker wrote several (...)

1“When a wise man points at the moon, the imbecile examines the finger”. Applied to sociology, the famous quote from Confucius could be translated as follows: “when readers1 are focused on the results, the major stake lies in the methods used to demonstrate them”. Howard Becker precisely dedicates his latest book to this very issue, that is to say under what conditions any data collected by sociologists can be accepted as relevant evidence supporting a general idea – what is generally referred as a “theory” or a “concept”. At nearly 90 years old, the author of Outsiders2 thus adds a new brick to his already impressive editorial (but not moral!) enterprise consisting of a reflection on the “tricks of the trade”3 of sociologists, while explaining them with a remarkable pedagogy4. In Evidence, he focuses on the different sources of the errors which can occur during the different steps of sociological work. The book is divided into two parts. Entitled “What it’s all about: data, evidence, and ideas”, the first part develops the meaning of each component of this scientific “Holy Trinity”. To begin with, Howard Becker builds on the works of the late Alain Desrosières concerning the development of models of inquiry in social as well as natural sciences, that is to say how researchers ended up in using statistics to objectify phenomena and produce “things that hold”. In the same chapter and still following Desrosières, Becker goes on to explain the opposition between the naturalists Linné and Buffon concerning the making of classifications, the former considering he could select a priori the relevant characteristics from among a given population, while the later rejected such an arbitrary method and preferred comparing species two by two to discover how they could be gathered or not. This radical difference of views is indeed still operative today in social sciences. In contrast to this Howard Becker argues that the opposition between quantitative and qualitative methods is nowadays outdated, summarizing the famous quarrel between Herbert Blumer and Samuel Stouffer during the American Sociological Association annual meeting in 1947, concluding that “The division of labor is more interesting than that. The different jobs [i.e. quantitative and ethnographic methods] interpenetrate, and a smart researcher does what needs to be done when the time is right” (p. 55).

2Before scrutinizing the really relevant methodological issues in social sciences, Becker makes a detour through the natural sciences and develops two examples: the first in physics, where a team led by Sébastien Balibar faced a perturbation problem involving radio waves during an experiment aiming to reach absolute zero; the second, borrowed from Bruno Latour, consists in decomposing the work of soil scientists determined to establish whether the jungle was encroaching on the savannah or the savannah was encroaching on the jungle in a Brazilian jungle.

  • 5 Art worlds.
  • 6 What Becker defined as a “collection of shared ideas and activities students used to organize their (...)
  • 7 Howard S. Becker and Blanche Geer, Everett C. Hughes and Anselm L. Strauss, Boys in white. Student (...)

3Entitled “Who Collects the Data and How Do They Do It?”, the second part of Evidence begins with a reminder that sociology is a collective activity like art, as Becker himself has already demonstrated5. Although sociologists are indeed eager to investigate the division of labour in many areas of activity, they tend to ignore or conceal this aspect of their own work. However, taking account of the fact that various actors with different motivations intervene in data collection makes it possible to identify some errors in this process. Chapter four is dedicated to the bigger social survey, the National Census. Becker highlights the difficulties posed by the mere counting of the inhabitants in a given territory and goes on to present the ingenious device designed by Peter Rossi and his team in the 1980s to establish the number of people who don’t “live someplace” in Chicago. Becker then questions the accuracy of the data collected in such surveys, developing the issue of ethnic classification in the US, alongside other similar problematic categories: religion or marital status. In chapter five, Becker focuses on the other types of statistical data gathered by Government employees. He thus presents classical issues about the measurement of phenomena such as suicide, homicide or crime and concludes by pointing out that “police statistics may present very unreliable information about the crimes they allegedly enumerate, but they’re very good information about some aspects of police activity” (p. 139). The next Chapter deals with other non-scientist data gatherers: the people reporting information about themselves and the “hired hands” who are generally not involved in data analysis and are generally paid according to the number of forms they fill in. This very aspect may indeed motivate them not to follow all instructions and then generate artefacts, as happened in one survey, incorrectly finding a rise in social isolation in the US between 1985 and 2004. In chapter seven, Becker uses other examples of surveys to show the various ways one can collect one’s own data, as many PhD candidates actually do. He also comes back to the irrelevance of the opposition between quantitative and qualitative methods, showing how research can benefit from their combination. This is precisely how Julius Roth demonstrated the magical rituals existing among medical staff, Donald Roy the workers’ self-restriction of production, or Blanche Geer and Becker himself the existence of a “student culture”6 while observing would-be physicians at Kansas University in the mid-1950s7. Conducting intensive field-work does not, however, prevent one from making mistakes. Chapter eight is dedicated to the different inaccuracies that may occur during qualitative research: such as forgetting historical change and writing reports in an eternal ethnographic present; ignoring specific conditions that may affect the object of interest; or making overly ambitious generalizations from the terrain whichs you have thoroughly explored. In such matters, one should indeed follow the footsteps of Everett Hughes – Becker’s “master” in Chicago – when he highlighted the ethnic division of labour in a small Quebec town while at the same time maintaining a certain humility humble as to the variables and processes that have been identified and allowing that they might take different values and forms in other contexts. Becker goes on to observe that: “What might be generalizable [are actually] the processes and subprocesses whose description researchers would have to adjust as they learn more about more cases” (p. 205), thus expressing clearly his agreement with Buffon rather than with Linné.

4All in all, with Evidence, Becker has written another must-read book for students and experienced sociologists, as well as also for all of those who do not yet understand how and why the sociology is a real science. Even though he does not obviously deal with all the methodological issues, he delivers another very tangible and lively account of what sociologists actually do when they work, and what they could do to improve their research. One might begin by meditating on his final word of advice: “Beware of traps and turn them into research topics”. Here lies the moon the wise sociologist points at.

Haut de page

Notes

1 It would be more accurate to say « publishers » insofar as many of them are used to asking researchers to remove most of the methodological aspects from the sociological books, or at least to relegate them to appendices.

2 Howard S. Becker, Outsiders: studies in the sociology of deviance, New York, The Free Press of Glencoe, 1963 translated into French by Jean-Pierre Briand and Jean-Michel Chapoulie (Outsiders, études de sociologie de la déviance, Paris, Métailié, 1985).

3 Les ficelles du métier, La Découverte, Paris, 2002.

4 Although his work as a whole contains strong methodological considerations, H. Becker wrote several books specially devoted to such issues: Writing for social scientists. (Chicago, University of Chicago Press, 1986), Tricks of the trade, op.cit., Telling about society. (Chicago, University of Chicago Press, 2007) and What about Mozart? What about murder? (Chicago: University of Chicago Press, 2015), reviewed on Lectures: https://lectures.revues.org/15949.

5 Art worlds.

6 What Becker defined as a “collection of shared ideas and activities students used to organize their response to the problems medical schools created for them” (p. 183).

7 Howard S. Becker and Blanche Geer, Everett C. Hughes and Anselm L. Strauss, Boys in white. Student culture in medical school, Chicago, University of Chicago Press, 1961.

Haut de page

Pour citer cet article

Référence électronique

Igor Martinache, « Howard Becker, Evidence », Lectures [En ligne], Les comptes rendus, mis en ligne le 08 novembre 2017, consulté le 28 mars 2024. URL : http://journals.openedition.org/lectures/23709 ; DOI : https://doi.org/10.4000/lectures.23709

Haut de page

Rédacteur

Igor Martinache

Professeur agrégé de sciences économiques et sociales à l’Université de Lille 1.

Articles du même rédacteur

Haut de page

Droits d’auteur

Le texte et les autres éléments (illustrations, fichiers annexes importés), sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search