[The final installment of the my current draft (this one positivist-in-orientation) on epistemology oriented toward the anarchist movement.]
Once the problems of defining anarchism and anarchist movements are solved (or at least addressed) and the question of research frame is dealt with, still more challenges await. These challenges are practical matters than restrict scholar access to anarchist movements, such as sampling, trust, and confidentiality.
In order to study anarchists or anarchism, one needs to find anarchists. Because overt anarchists are relatively small in number today, it is tough to locate anarchist populations to sample. Covert anarchists obviously do not wish to be found and almost surely seek to avoid being studied. Where there are important ethical concerns to consider when studying covert anarchists, such research is still meaningful, valuable, and worth the effort to protect privacy. The behind closed-doors decision-making of covert anarchists can help to inform the future efforts of anarchists, who act under the gaze of watchful states and other authority figures. For example, what strategies are used for affinity group formation? How do covert anarchists participate overtly in the mainstream? How do people decide to engage in illegal activities instead of legal activities?
The fact that covert anarchists wish to avoid study is relevant within representative democracies, where information on membership in affinity groups, Animal Liberation Front cells, or squat residences are completely unavailable. But, this reluctance is particularly true in authoritarian countries. For example, the Nigerian anarchist organization called the Awareness League requests that anarchist comrades not include the “Awareness League” name on mailed letters (or any mention of political matters) out of fear of government censorship. Likewise, anarchists in Zimbabwe fear government repression (with good cause!), and consequently conceal their politics behind “arts and culture” community groups. More generally, locating anarchists requires some sort of personal contact information that most people do not regularly have. Consequently, having access to broad social networks is especially important for those interested in studying anarchist movements.
Even when located, anarchists are often hesitant to help scholars pursue their research. Why? Anarchist movements have long been spied upon and then persecuted by governments, corporations and wealthy land-owners, religious institutions, and others. Anarchists are usually aware of this history, and when combined with present-day repression, activists may develop a keen sense of paranoia. Thus, access to movement participants may be thwarted by distrust of outsiders, including (and maybe especially) academics. North American anarchists sometimes call this healthy paranoia “security culture”, which results in a refusal to speak about tactical matters (and sometimes political matters) to anyone they do not know personally and well. Inter-organizational matters of dissension and controversy may be particularly off-topic to outsiders. As many scholars are employed at publicly-funded universities, scholars may be distrusted as either interlopers with little sympathy for anarchist goals or as spies for governments. The former are viewed as parasites who make careers off the struggles of others, the latter as enemy agents intent upon disruption.
Anarchist activists have other reasons to be skeptical of participating in academic research. Even for studies conducted by activist-scholars with the best of intentions, the research process is fraught with risks. Confidentiality may be threatened by authority figures, especially if research seems to isolate and identify individuals. What is to stop a powerful elite from trying to track down original contacts, interviewees, or survey-takers, in order to spy on them, arrest them, or worse? Or more mundanely, employers could harass or fire, landlords refuse to rent to, and family and friends could shun if anarchists' identities became public in the wrong way. Even when confidentiality has been actively protected, the very existence of published research may be a valuable tool to attack a movement. Research that helps scholars and activists to better understand how movements work and how to improve strategies, can also be used by the state to identify movement weaknesses and target those movements for disruption, especially if source data become “public data”, which anyone could then access and analyze independently. Anarchists are rightful suspicious and hesitant to help authority figures undermine their own movements.
Anarchist-sociologists will likely have to go out and collect such data, participate in the aforementioned types of conversations, and live and observe these experiences. For those with no immediate access to research subjects, scholars might wish to seek out other forms of data to analyze. Unfortunately, there are no real good sources of data on anarchists for obvious and previously discussed reasons. There have been some secondary sources we have found useful, albeit imperfect. Very few surveys have been done of anarchists, although some data exists on attitudes about anarchists. One of us has published research using online surveys (e.g. user survey from Infoshop.org), which while not too generalizable, can answer some questions about individual anarchists (see Williams 2009a, 2009b). Organizational directories, such as the International Blacklist (from the 1980s) or the more recent Anarchist Yellow Pages, catalog the existence, location, and character of anarchist organizations (see Williams & Lee 2008). Recent anarchist history has been and is being compiled by various projects, like the A-Infos News Service (see Williams & Lee forthcoming as a use of this “data”) and the Independent Media Center network. These projects—and many others—describe and record anarchist events, political news, organizational and movement histories, and other documents of note.
Mainstream and alternative press could be used as data for content analysis that indicates the popularity, controversy, or successes and victories of the anarchist movement. For example, long-running newspapers like the New York Times has recorded decades of anarchist history—albeit through an unfavorable, elite lens and in superficial contours. Library collections—such as the Labadie at the University of Michigan—have many complete series of anarchist newspapers and magazines from the near and distant past (Herrada 2007). One underused—and perhaps less accessible—source of data is that collected by various law enforcement authorities, who often have a near-pathological institutional mandate to collect information on enemies of the state (see Borum & Tilby 2005 for analysis premised upon police data and perspectives). The data law enforcement collects will generally reflect the perceived, relative threat posed by anarchists and anarchist movements. Still, this kind of data could be a fascinating treasure trove.
The preceding sources of data are not only suggestive of possibility, but also deserving of caution. We end this chapter with a brief discussion of methodological strategies and their weaknesses as they pertain to the study of anarchist movements. The major concerns that deserve attention include incompleteness, inaccuracy, non-generalizability, and challenges to access.
Standard sociological methods, such as survey questionnaires, provide some nice benefits, but are rife with problems when studying anarchists. For example, it is impossible to properly sample anarchists, since no sample frame for anarchists exists. Compared to other groups of people—maybe Sierra Club members who are recorded on master lists by the national organization—it is impossible to know who all the potential anarchists in a given area are. Therefore, any anarchists that are included in a study will not accurately represent the entire population, since the constitution of the entire population is unknown. Additionally, an unavoidable selection bias results since not everyone given a survey will actually complete it—consequently, any conclusions drawn from these surveys will only reflect the patterns of those who choose to do the surveys, not those who refuse them. Thus, surveys are bound to be non-generalizable to the larger population of anarchists. Even efforts to ascertain opinions about anarchists, collected by generalizable surveys, will still be problematic, as secondary data can rarely answers questions that they were not designed to answer originally. Researchers will be grappling with operationalization issues and trying to apply old data to new research questions.
All manner of field methods—ranging from ethnographies to simple face-to-face, open-ended interviews—have other, unique challenge. One formidable problems, previously discussed, has to do with the establishment of trust between anarchists and those wanting to know about anarchism. As a result of distrust of outsiders, committed anarchists are probably the best people to interview other anarchists. As with all qualitative research, questions of generalizability abound: are the people who get observed or interviewed at all alike other anarchists? It is likely that those willing to be observed or interviewed have certain commonalities—e.g. an outward focus, overt status, etc.—that set them apart from other groups of anarchists, such as those who participate in black blocs, do targeted property destruction, advocate armed struggle, are racial or ethnic minorities (or of immigrant status), have been previously arrested, and so forth. There is also always the question of time and expense when doing in-depth observation and interviews: such research tends to span long periods of time and require substantial financial resources for travel and living expenses.
Passive methodologies that do not require direct contact with research subjects, like content analysis, introduce still different issues. Depending on the medium under study, content analysis could be biased (in one direction or another), incomplete, or simply unhelpful. In each case, it is important to honesty consider the source of the data under analysis and reflect on just how helpful it can be. For example, movement documents—statements from anarchist organizations, the movement's magazines, or claims made in anarchist histories—may reflect desires and critiques of anarchists, as opposed to reality and fact. Thus, all movements (anarchism included) put forth propaganda to convince allies to join them, enemies to give up, and for themselves to keep fighting. It is likely quite regular for movements to brag and exaggerate about the success of actions or the attendance at an event in order to make it seem better than it actually was. Movements (and obviously the groupings and organizations within them) are not always good at keeping records about themselves; their decisions, the logic that led to those decisions, and the result of decisions are regularly omitted in meeting minutes or public pronouncements. Thus, for researchers trying to understand how and why something was done, movement documents might make it seem as if nothing actually happened, although the opposite is more likely true. Consequently, many researchers have to read between the lines, as troubling endeavor as it requires projecting one's own conclusions about other's actions.
The mainstream media is an equally complex source of data for content analysis, but for different reasons. Media often present anarchists in a negative light—when they are presented at all. Thus, an “anti-halo effect” likely surrounds anarchist activities. Selective coverage will focus on the most dramatic (and probably most unflattering) behaviors, such as property destruction, while mundane, heroic, or positive acts like sharing food with the homeless, watching police patrol poor neighborhoods, or making long, consensus-decisions at meetings will be overlooked. The the media includes a range of anarchist behaviors that is non-representative of the true spectrum of activities. Last, media regularly misquotes people or places their words and actions out of context—whether through reporter misunderstanding, deadline pressures, or deliberate, ideological editorialization. While, this problem affects many movements, it is bound to be of particular significance for radical movements like anarchism who have ideas beyond the typical realm of discourse and otherwise unintelligible to mainstream media.
Finally, going further into protectors of the status quo, one could analyze records kept by law enforcement. While, there are likely extreme issues of limited access (as mentioned earlier), there are also problems related to police mischaracterization. Thus, something that has a clear meaning to anarchists, may be highly misunderstood, miscategorized, and presented in a fashion that misconstrues anarchist intentions. Also, given police reliance upon informants (paid or unpaid), researchers would have to consider the motivations of those relaying information. Do informants seek to validate their own importance to police by playing up perceived anarchist threats—or even to insure the continue to get paid for their spying? Or, are informants telling police what they thing police want to hear? And if so, what exactly is that “truth”? Regardless, content analysis of law enforcement records would have to sharply filter data based on these rampant biases. Such problems should not deter the use of this kind of data—in fact, it may reveal a good deal about how police perceive and frame threats posed by anarchists—but it should caution the conclusions arrived at.
 Not coincidentally, paranoia is one of the very mental states that governments seek to foster amongst radical movements.
 Ironically, anarchists have been unintentionally instrumental in the founding of the FBI and Secret Service in the US (via FBI founder J. Edgar Hoover's pathological obsession with “reds” and his participation in the deportation of anarchists Emma Goldman and Alexander Berkman to Russia in 1919 and the backlash following Leon Czolgosz's assassination of President William McKinley in 1901), and the European-based Interpol policing agencies (Jensen 1981).
Borum, Randy and Chuck Tilby. 2005. “Anarchist Direct Actions: A Challenge for Law Enforcement”. Studies in Conflict & Terrorism, 28 (3): 201-223.
Herrada, Julie. 2007. “Collecting Anarchy: Continuing the Legacy of the Joseph A. Labadie Collection”. RBM: A Journal of Rare Books, Manuscripts, & Cultural Heritage, 8 (2), Fall: 133-140.
Jensen, Richard Bach. 1981. “The International Anti-Anarchist Conference of 1898 and the Origins of Interpol”. Journal of Contemporary History, 16 (2), April: 323-347.
Williams, Dana. 2009a. “Red vs. Green: Regional Variation of Anarchist Ideology in the United States”. Journal of Political Ideologies, 14 (2), June: 189-210.
Williams, Dana. 2009b. “Anarchists and Labor Unions: An Analysis Using New Social Movement Theories”. WorkingUSA: The Journal of Labor & Society, 12 (3), September: 337-354.
Williams, Dana and Matthew T. Lee. 2008. “'We Are Everywhere': An Ecological Analysis of Organizations in the Anarchist Yellow Pages”. Humanity & Society, 32, February: 45-70.
Williams, Dana and Matthew T. Lee. Forthcoming. “Overthrowing the State Without Using It: Political Opportunities for the Anarchist Movement”.