Saturday, 19 November 2011

On the Very Idea of Security and Terrorism Research

A little while ago I posted about a debate in my country with regard to research on and press reports about alleged former Swedish agents of STASI – the security agency of the former German Democratic Republic (GDR). The debate continues, so far with rather indeterminate results, but the issues raised by Birgitta Almgren's book (see the former post) got me thinking about issues in other areas that connect to a wider set of questions with regard to the very idea of conducting proper research about secret or security related activities, people and organisations. I touched on this issue in my former post when pointing out that, from the point of view of the well-tempered scientist, empirical research where data is not readily available for independent inspection by others is a rather odd – if not self-defeating – concept. So, in effect, can there really be such a thing as "security research" and "security experts"?

This question is interesting not only from an academic point of view, but just as much – if not more – from a general societal one. Nowadays, whenever there is some significant turmoil or development in international politics, such as riots, concentrated propaganda or violent attacks seemingly targeting a well-defined group, semi-military strikes of the form that we often like to call terrorism, and so on, you can bet a few months salary on there being a stampede of people in the media labelled "security researcher", "security expert" or something of the kind. Sometimes the word "security" is exchanged for "military", terrorism" or "intelligence", but that makes no difference. Sometimes, the members of the herd are attached to public research institutions, sometimes to governments, sometimes to what has become known as "think tanks" (i.e. interest organisations posing as public research institutions). But that doesn't matter either – in the present context at least. What does matter, though, is that they all are being held out – and normally hold out themselves – to possess special insight, advanced knowledge or expertise worthy of special trust. It is because of this that these people get to shape our perceptions and views of world and domestic events through the media. It is because of this that they are employed by government agencies or called in to brief policy makers when the need arises. So, effectively, the question I will be asking is not only about to what extent, and on what conditions, proper security research can be done, but also to what extent the credibility apparently enjoyed by the sort of people just described can have any substance. Again, the basis for my doubts is the very nature of the object of the alleged research, insight or expertise – that fact that when the factual basis of the alleged research is secret, covert or otherwise inaccessible for others, the very idea of proper research is undermined in a fundamental way.

Why is this particular feature of research – open access to data, research protocols, and so on – so important? Well, one of the things that is usually held out by philosophers and theorists of science is that science and research are essentially a collective enterprises. While in the eyes of the public and mass media, a research result is delivered when an article, report or book is published, from the point of view of the research community in the field in question, this is merely the first step towards a result. The result is eventually produced through a complex interaction of checking for mistakes and misreadings at different levels, offering alternative interpretations or explanations, questioning the validity of arguments, and so on. This goes on in all areas of research, regardless of whether it is qualitative or quantitative or both, whether its main questions are truth oriented or exploratory interpretative or both, and so on.

A nice example from the biomedical and natural sciences regards the breakthrough of the first reproductive cloning of a larger animal. The cloning of the sheep Dolly was first reported in a letter to Nature in february, 1997. Many believe that this is the time when the possibility of this sort of cloning was discovered. However, in the scientific world this was very far from the truth. Although Ian Wilmut and his team have been retrospectively credited for delivering the first report of the first successful attempt, the discovery that this was in fact the case was made almost one and a half years later, after a meticulous checking and re-checking of data, possible mistakes and alternative explanatory hypotheses undertaken and eventually evaluated by many people outside of Wilmut's group reviewing the data and trying the methods on other animals (some sources of the reports of this work are here, here and here). That is when a research result – rather than a report and a conjecture – came into being and the producer of this result was the collective of all of the people involved throughout the process. A more recent example of a similar thing can be seen in the example of the surprising report from physicists at CERN that they have measured particle speeds higher than that of light. In the very same report it is said "independent measurements are needed before the effect can either be refuted or firmly established". In other words, this is a report and a conjecture, whether or not it is a result, time and a probably very heavy load of work of many other people besides the CERN team will tell.

Now, this was natural science and it may be believed that things are different in at least some disciplines in the humanities or social sciences. This, however, would be a misconception. Indeed, the humanities and social sciences to some extent deal with very different research questions than the natural sciences and, for sure, their methods are to a large extent rather different. Indeed, some areas of the humanities (such as certain areas of  literature) do not even make inquiries where evaluations in terms of true/false are very relevant (although this should not be overstated, since also in these areas claims are being made and defended – e.g. about an interpretation of a poem being possible, interesting or innovative). But none of these special features takes away the fact that also in these field, the achievement of research results is essentially a collective matter. Just as in the natural or biomedical sciences, when articles and books make conjectures and report work done in the humanities and social sciences, this is just the first step towards a result that also in these fields in the end emanates out of the combined efforts of many. And, in the humanities in particular, this is normally evident regarding the work leading up to publication – a simple proof being the extensive acknowledgement sections normally attached as a preamble to publications in this field.

Now, to have such a collective process of criticism and improvement, access to data, details about method, sources and so on is, of course, of vital importance. If that cannot be had, the process cannot develop according to the standards of research and, consequently, nothing deserving the label of research has occurred and, even more so, no research results will ever come out, regardless of how much the investigator in question applies methods, tools and theories commonly used in research.

So, what about security research, then? Well, it is obvious that some research with regard to secret political or military operations is indeed possible. This regards mostly historical research looking at events from a rather long time ago, where archives have been opened and seals of secrecy lifted. But even then there may be severe limitations, e.g. due to incompleteness of the archives (sometimes due to deliberate cleansing) or inaccessibility of what spies call "humint", that is informants who personally experienced or otherwise possess credible knowledge about what documents and records describe. Moreover, when the research regards what is effectively spy activity, to have a lot of humint from different angles is essential – and this for the very same reason why humint is essential in spying and intelligence work: Whatever there is on paper (or harddrives) may in this area of reality be interpreted in very many and mutually exclusive ways.

Say, for instance, that you find genuine documents from an intelligence agency of a country, A, listing someone from another country, B, as an agent or a friendly source. Call this person Max. Does this mean that there are good reason to believe that Max indeed was spying for A? Not necessarily, the documents may be assessment reports reflecting mere perceptions of an officer in the intelligence services of A – perceptions that may have been dismissed. But suppose that we find more documents pointing in the same direction. Not only Max's name on a list, but also summaries of reports allegedly supplied by Max, assessment by independent officers and the receivers of the intelligence allegedly coming from Max, details of operational procedures to be applied with regard to Max, and so on. Isn't this quite strong proof that Max was spying for A? Not at all, and this is clearly seen when we ponder the obvious fact that, for an intelligence organisation, there are many reason for storing documents in its archives besides keeping track of actual events. The most obvious reason would be to authenticate what in old KGB terminology was called maskirovka, that is, operations aiming to deceive rather than collect knowledge. There are many well-known military examples of this, but in peace-time, spreading disinformation is mostly about either influencing political development in a country or having one's counterpart in another country set off on a false trail. This may, for instance, be about discrediting a person assessed to be a valuable asset for the other country. So, the intelligence or security service of A may have viewed Max as such a person with regard to B and thus developed plans to undermine his career. Under the prudent and quite realistic assumption that B would have sources within the intelligence and security community of A, such a plan cannot be put into operation without the discrediting information about Max being possible to authenticate through the presence of actual documentation describing Max's treason against B in the archives of the relevant organisations of A. Another possibility is that A wants to be able to point the finger at Max in order to deflect suspicion from another, real, spy or friendly asset in B. And so on. In a well-run organisation practicing maskirovka, many such false trails would be present in its archives, several of which devised for various eventualities, neither of which may ever be actualised.

It is difficult not to come off as a conspiracy theory nutter when describing possibilities like these. However, if you propose to be doing research, say, on the operations of security and intelligence involving A and B, such possibilities cannot be overlooked. After all, research is supposed to be something more advanced than just common sense thinking, the knowledge produced by it is supposed to be especially credible and far-reaching. So, minimally, then, it would be necessary to check what the presence of the documents pointing the finger at Max is due to according to some people involved. One of these would, of course, be Max, colleagues of Max, friends and family of Max. Others would be relevant officers of the organisation of A alleged in the documents to be running Max as an agent, as well as their counterintelligence and intelligence counterparts in B. And so on.Yet another thing to check is if there are sources of information (people or documents) that have not yet been checked and that might cast light on the matter. In addition, for each of these possible sources, assessment has to be made with regard to credibility and their possible reasons for distorting the truth or the possibility that they in turn have been deceived. And so on and so forth. The trouble, of course, is that none of this can usually be accomplished. Either because the people have vanished or died or don't want to talk, or because relevant documentation is withheld, lost or destroyed. Or, which is very common in the case of the so-called terrorist experts, sources are kept a secret – leaving it a very open question whether or not there are any sources in the first place.

Take the case if the recent research into STASI operations in Sweden. This research is based on a fragment of documentation held by the Swedish security service. In addition, Almgren consulted the STASI archives in Germany and interviewed a few people there. However, as she recently pointed out, there are many more documents believed to be held by the CIA, for whatever reasons that CIA see fit. Also, we can be certain that the German security service has gone through what there is and made its selection. Moreover, the documents there are are those left behind by officers of STASI in the first place – the base of archival evidence in this matter thus having been originally shaped through choices made by those people, for whatever agenda that directed their actions at the time. On top of that, we have the far from improbable possibility of there being lots of documentation, the existence of which is withheld from the eye of researchers by any of these parties. In addition to this, Swedish law prohibits Almgren from making the identity of the people she describe known, or to approach them herself for interview purposes. Neither can any other researcher, by implication. In all, this makes a strong case for the claim that the prospect of having any research results in this area is close to nil. This is not changed by the fact that Almgren is an esteemed researcher or that she has done as well as she could, using the tools of the research trade available to her. In the end, the outcome will never be more than an (admittedly unusually advanced) piece of journalism. My own personal opinion is that she should therefore not have published the book, since under present circumstances it is just as likely to distort the truth as making it clear, while at the same time causing human suffering. Instead, she should have done from the beginning what she is doing now: calling for  changes of the circumstances so that real research becomes possible.

Now, in the case of Almgren, at least there is an honest researcher doing her best under the circumstances and disclosing what she can disclose (although, perhaps, she shouldn't have). With the various security or terrorism experts appearing in the media as soon as there is a bombing, a shooting or any other sort of semi-military violent event or conflict the story is much, much worse.

First, many of these people are employed by actual players in the very games on which they are called to give expert input – intelligence or security services, foreign affairs ministries, the military, political party affiliated organisations, and so on. All of these have their own agendas in the matters on which the alleged experts are called in to testify. The worst case of all are the so-called think-tanks, where the actual agenda, funding sources and mission-setters are as a rule carefully hidden from the public behind a marshmallow-ish cloud of noncommittal declarations on a homepage. Now, had they actually been real researchers, this may not have been so serious, since anyone could then have checked what they are saying, their evidence, their sources, and so on. But, of course, this is as a rule impossible. These people at best come out as spin doctors or sensation journalists attempting to instigate a smear campaign – referring to one unidentified but allegedly credible and well-placed source after another. Most of the time, however, it's not even that – it's just them and their alleged expertise. Tragicomical enough, this claim to credibility is often backed up by citing the very thing that should make any person doubt what they have to say – their connection to any of the sort of organisations mentioned above.

Now, most of these self-professed experts are smart, at least in a street-level sort of way. If you listen carefully to what they have to say the next time they're on the news, or read those of their reports that it is possible for a citizen to access, there isn't an awful lot of specific, controllable information in there. Whenever a sentence seems to make a claim of a sort that would be possible to fault or confirm, you will find it surrounded by qualifications draining it of all substantial content. What is left is the emotional and evaluative tendency: certain parties or people are being associated with certain emotionally or evaluatively loaded words or phrases. Observe, for instance, how these people – without a hint of an argument – shape and protect a certain conception of what events can be called acts of terrorism. So my own answer to the riddle of why on earth they are being employed by anybody is quite simple: the salary is coming out of the propaganda budget.

But don't they ever utter a testable conjecture, these people. Indeed, on scattered occasions they do. Let me take a recent example from my own backyard. I trust that you are familiar with the bombings and heinous killings in Oslo, Norway, committed by the tall, blond right-wing extremist Anders Behring Breivik. When the news about the bombings broke, it was immediately commented on by my country's most notorious so-called security and terrorism experts, Magnus Ranstorp – in the very words on his homepage at the Swedish National Defence College, where he is employed:

Over the last 20 years Dr. Magnus Ranstorp has worked closely on terrorism and counterterrorism issues worldwide with a special focus on Islamist extremism
Ranstorp's initial message was that the Oslo and Utøya mayhem and tragedy was most probably an attack by Al-Quaeda (here, and here). This conjecture was based on no evidence whatsoever besides the credibility of Ranstorp's self-professed expertise. A few hours later, when it had been confirmed that the attacker was a tall, blond Norwegian national, Ranstorp once again sought publicity, now saying in Sweden's largest daily that this was "unprecedented" and quite beside the dominant trends (here), this in spite of the fact that nationalist, separatist and right-wing extremist (Breivik's motives seem to be a mix of these) motivated violence has been dominant during a rather long time in official European statistics – in particular if the UK is seen as a special case (see the latest report on European terrorism and trend from Europol). It may be added that Ranstorp is on record as viewing the Europol statistics, the closest we have to any sort of verifiable research data in this area, as being a political tool for deflecting what he is convinced (naturally for undisclosed reasons) is the largest and most important threat: islamist terror-groups. A few days on, when Breivik's motives and planning where starting to become known, Ranstorp raced to the fore once again, now conjecturing that even if Breivik was anti-Islam and anti-Islamism, he was at the root a product of Islamist terror (in Swedish media described by Ranstorp as a "hybrid", actually a little bit of an Islamist besides the anti-islamist motivation), since he had learned a lot of nouts and bolts about bombing and terror tactics from them (here, for once in English). I suppose that Ranstorp would be quite willing to generalise the principle thus applied, in consequence claiming Al-Quaeda terror to be partly Zionist (as practiced before the creation of the current state of Israel), this Zionist terror being partly Nazi, Nazi terror as practiced by the SA in the 1930's as partly Catholic (the pogroms during centuries), and so on and so forth. Wonderful to have a sharp mind at work in the service of mankind, isn't it? In addition, in all of these rants, Ranstorp is silently nudging the reader towards the notion that, in fact, Breivik is not even a terrorist, he's a nutter - apparently implying, in contradiction to what he says in general about right-wing terrorists (see links above),  that terrorists by definition have to be quite sane people. So much for the notion of researchers and experts on terror and security.

But behind this comedy, there are more serious issues to consider. Ranstorp is quite obviously a propaganda megaphone posing as a researcher. The only explanation for keeping him employed at the National Defence College appears to be the one offered above. His main agenda, apparently, is to preach the teaching that terror worthy of the attention of policy makers and security organisations is islamist terrorism. When facts don't suit Dr. Ranstorp, he declares them to lack credibility (they have to, don't they?!) or bends them in the dodger manner exemplified above. But Ranstorp is not alone. He has colleagues, who profess the same dogma in more clever ways, not sticking out their necks with propositions that are in fact open for straightforward scrutiny. Magnus Norell is my country's other notorious security and terror expert, employed at the Swedish Defence Research Agency. In the same context where Ranstorp declared that he didn't have to pay attention to the Europol statistics, Norell made the more cautious claim (quite consistent with named statistics) that the real problem is islamist ideological "radicalisation", which he claims "can lead to acts of terror" (here). No shit, Sherlock. But, of course, Norell's message isn't as trivial as that. The message is that we have reason to, at least tentatively, believe that there is a causal or explanatory link between embracing an ideology of the type that Norell refers to and participating in politically motivated and organised acts of violence targeting civilians. Norell, in contrast to Ranstorp, seems to be a reflecting person, so I suppose that he is aware of the sort of evidence that would be needed to support a conjecture like that. For instance, one would need to compare an abundance of currently non-existing statistics on the varying ideological leanings (or lack of such) of people and their actions at later times. This evidence is simply not at hand and will not be in the foreseeable future. Being charitable, I also assume that Norell is familiar with the problems of distinguishing socialisation and selection mechanisms in social science research. That is, even if there was to be a statistical correlation between belief in the ideology in question and later acts of terrorism, this can just as well be explained by the suggestion that people who are prone to commit such acts are also prone to embrace beliefs that justify them to do what they desire to do. The prospect of checking for that, I conjecture, is once again close to nil. And this, I suppose, Norell knows very well, but nevertheless keeps on presenting his pet hypothesis as if there was in fact some sort of support in its favor. Norell, of course, presents none – besides his academic and employment credentials.

So, to finally return to my original question. Can there be such a thing as security research of the type discussed in this post? Well, in a way there can, if only the researchers in this field behave as serious researchers in relation to the dramatic limitations of the field. The main effort to be done in this area seems to be about collecting and securing universal access to reliable data, including routines for checking these against each other, thereby improving basic tools such as statistical analyses of trends, and so forth and making them usable in a real research process. This is probably pretty frustrating for many of the people currently involved in this area – they, as we have seen, want to know the answers to the big questions! Unfortunately, that knowledge is unattainable as long as they neglect to secure the basic building blocks of any sort of social science research. At the same time, the way that the field has organised itself seems to work against the prospect of this happening any time in the future. Funding is coming mainly from organisations, the main interest of which is to withhold, distort or fabricate information on the basis of a political agenda of national interests. And when this is not the case, the very same organisations decide on limitations that make actual research impossible.