Falsifiability, the Politics of Evidence, and the Importance of Narratives

,


Falsifiability, the Politics of Evidence, and the Importance of Narratives
One of the more famous definitions of the scientific method is that of falsification.This definition is found in the work of Karl Popper (2002, 22) who writes that, 'According to my proposal, what characterises the empirical method is its manner of exposing to falsification in every conceivable way, the system to be tested.'Regardless of whether or not Popper has the final word on what constitutes science, it is fair to say that this definition remains foundational to contemporary understandings of how scientific work is done.Indeed, it is not a definition that I am interested in contesting.Instead, I am going to suggest that this definition provides us with an opportunity to understand, first, how the rise of alternative facts (or debates about 'truthiness', as Stephen Colbert would have it) became possible and, second, why narratives are so important both in the construction of knowledge about the world, and for our ability to develop normative arguments as counters to alternative facts and a post-truth politics.This discussion is important for thinking about security because all claims pertaining to a security matter depend on some kind of evidence.There is no threat without believing that the threat is real.There is no security risk without there being a danger of some sort.There needs to be evidence justifying decisions that send soldiers into harm's way (hence the debates in the lead up to the Iraq War and Colin Powell's presentation to the United Nations).
The creation of fake news as a weapon to be deployed, as it appears Russia did by using Facebook and other media to influence the American presidential election, is further ground for taking very seriously how factual claims function as evidence in our public political and security discourse.The rise of alternative facts is not just something for reporters to be concerned about.It is of concern to scholars of politics, security and international relations.Those who challenge such unfounded claims are accused by those presenting the alternative facts of peddling fake news.The significance of Conway's invention of 'alternative facts' may not be especially surprising as politicians in general have a reputation of playing fast and loose with the truth in order to achieve their goals.Yet there is something that is if not new here, of particular significance because of the Trump White House is prepared to not only regularly dismiss evidence, but to do so by attacking evidence as fake, and make up their own evidence instead (often by tweeting).Trump's claims about the dangers posed by Mexicans and the need for a wall along the border is a further example of how political policy claims are being made without recourse to any evidence.It is not beyond the realm of possibility that he could provoke a trade war (if not a real war with North Korea) with countries he accuses (without any evidence) of giving the US a bad deal While this characterisation refers primarily to the Trump White House and its supporters, a similar pattern exists elsewhere.For example, in the UK during the referendum about whether or not to leave the European Union (EU), senior member of the Conservative Party, Michael Gove, dismissed the claims of the Remain campaign on evidentiary grounds when he claimed during the referendum that, 'People in this country have had enough of experts' (Mance 2016).Indeed, much of the debate during the referendum was over evidence, or rather the absence of it from the 'Leave' side, which clearly misrepresented the consequences of leaving the EU.It is very plausible that they were aware of the deception, hence the decision to scare people into voting Leave by focusing their campaign on societal and other security risks ostensibly posed by immigration (Watt 2016;Hall 2016).
Underlying the fake news position is the respectable logic of falsifiability, and of the ways in which facts are regularly presented to support opposing political positions.To those who do not understand how falsifiability works, it easily appears as though facts can be generated to support almost anything.Moreover, the public is not always able to interpret scientific findings nor understand how the scientific method works (nor appreciate that within academic training we learn to constantly question each other's conclusions, but that this does not necessarily diminish the strength of research findings), so when a study is challenged it lends credibility to the popular view that facts are only as good until the next one comes along that replaces or debunks the previous one.In this sense, the very idea of falsifiability offers a generalizable position whereby any fact and any knowledge claim is suspect.
The scientific counter to this position, that there are empirically objective claims, is unhelpful and also largely irrelevant.Simply put, it does not matter whether or not there is a superior empirical basis for any particular claim if the process by which the claim is made can be manipulated and if the claims themselves are regularly contested and occasionally refuted.
There is a paradox here whereby the process of producing facts creates the conditions for undermining these same facts.
Contemporary politics aside, the methodological issues at stake here are not new.It was on the related question of relativism that Alan Sokal (2008) launched his polemical attack against post-modern or post-structural social and political thought during the so-called science wars of the 1990s.His argument should concern anyone interested in normative work, since it was the ability to deploy evidence in support of normative positions that drove his critique.Then, as now, this natural-scientific position was flawed on its own terms because it ignores the political consequences of how falsifiability or related testable requirements of scientific research methods contribute to, and may even undermine, the meanings of scientific conclusions.
The production of new knowledge and the discovery of new facts do yield advancements in knowledge.However, so long as our focus is exclusively on searching for or producing more accurate facts, it is easy to undermine other factsless accurate onesthat do not measure up to the required standard.This constant undermining of previous knowledge may be the norm for academics, but in the public sphere it can be reframed from being part of the scientific method to undermining the claims of academic research.
Moreover, much like the deployment of the 'fake news' label for political purposes, methodological claims can be used within academia to advance personal ambition at the expense of others.The story of Amy Cuddy, recently reported in the New York Times Magazine (Dominus 2017), is indicative.Dr Cuddy became famous for her psychological work about 'power poses' that suggested that certain physical poses can yield an increased sense of confidence along with physiological changes in testosterone and cortisol levels.Her work, however, became a target and was vigorously attacked by other scholars in the field using different statistical tools as their weapon (although ego, personal ambition, and gender were also involved).Hers is a striking example of how a debate about method (in this case, about p-values) was used to shatter a scholar's academic career and promote that of her attackers, without considering the greater (and obvious) methodological point that all claims to specific interpretation over methods are equally logically falsifiable as any other. 1The kind of intense 'methodism' (Wolin 1969) that was used to attack Cuddy is disingenuous in how it ignores the clearly obvious interpretive (and power) dynamics that inform how methods are applied and deployed in a professional capacity.The extent that scientific work is characterized by different methods and multiple interpretations of how to apply specific methods is partly why Paul Feyerabend (1975Feyerabend ( (2010))) argued that there is no single method that defines science.None of the above is to suggest that there is no such thing as evidence.Indeed, evidence matters a great deal, especially in the construction of normative argument.Questions about 1 In science, this debate is more easily witnessed by looking at the problems inherent in induction.A good survey of this and other methodological debates of relevance is (Chalmers 1999) insecurity and injustice often rely on evidence to demonstrate their veracity.Moreover, explanatory and normative work both invoke causal claims without necessarily asking what kind of causality is being invoked.2However, the focus on the production and contestation of facts and evidence can serve like a smokescreen, directing attention towards the grounds on which evidence is basedoccasionally leading toward ontological claims as critical realists do3 when the point of debate ought to be about the hermeneutic framework in which the interpretations are made.In research areas that speak to the socially constituted and socially constructed world of our interactions, the form of the interpretations in which we locate and debate facts are narratives. 4olitical theorists have noted the centrality of narratives in normative work (Crary 2007;Schiff 2014).Richard Rorty (1989, 108) goes so far as to say that narrative is what the theorist does: 'How can one be a theoristwrite a narrative of ideas rather than peoplewhich does not pretend to a sublimity which one's own narrative rules out?' Narratives provide a story in which we are able to locate claims about the world, ourselves, and others, and interpret their meaning.Through narratives that make sense to us, we find the hermeneutic framework that transforms the world into sensical meanings.Facts, consequently, are always produced in and by a claim to a particular narrative.This claim does not mean that everything is relative because the narratives are meaningless if they cannot be shared in a meaningful way with others.As Anna Wibben (2011, 43) notes, we are not only beings that create meaning (homo significans), but are also 'homo fabulans because we interepret and tell stories about our experiences, about who we are or want to be and what we believe.'Wibbens argues that security studies needs to take narrative seriously, and methodologically this means turning toward ontology and the related methodology of interpretation.
Those who rely on ontological claims to ground what constitutes facts are right about at least one thing: we need to take ontology seriously.However, they are wrong to dismiss one of the central insights of an ontologically-focused philosophy, which is that interpretations are a central part of our being.This insight is why Martin Heidegger (1962Heidegger ( (1999))) refers to his methodology as hermeneutic-phenomenology.However, one does not need to turn to Heidegger to find that interpretation matters as a significant aspect of our being in the world within which we find ourselves.
Relatedly, Max Horkheimer (1975) noted in the 1930s that there needs to be a distinction between the social and political contexts in which we are able to produce a type of knowledge that can be deployed as a fact, and a methodological awareness about how we value different types of knowledge.He identified that research cannot be divorced from the conditions in which it is produced.What can be loosely described as the Continental tradition of philosophy and political thought has repeatedly offered multiple avenues by which to recognize the centrality of interpretation, and thus also of narratives,5 for our political being.
While not all interpretive activities can be described by a reference to narratives, narratives can be understood as the framework in which our interpretations exist.People filter knowledge, and no pretending about some objective universe of facts can overcome this characteristic of human beings.This act of filtering is the means by which we identify what claims make sense, and how, but it functions by a recourse to our own self-understanding.
The post-Kantian tradition has tried to resolve how this process of making intelligible functions through particular epistemic categories.What this line of thinking does, however, is remove our own construction as beings from the act of knowledge production.One of the achievements of phenomenology is locating ourselves centrally in the production of meaning.
By emphasizing that our existence as particular beings in specific contexts is important in how we interpret meaning in the world, identity takes on a greater role.It is, for this reason, unsurprising that public political debate often turns on identity claims instead of ostensibly objective factual ones.6Indeed, narratives cannot make sense if we are unable to locate ourselves in them in some way, and thus they always pertain to identity-claims.The implications of how this form of interpretation works for security is evident in how we navigate competing interpretations of the same facts.
For example, in the case of Black Lives Matter versus Blue Lives Matter, 7 the facts are in a sense the same: the number of deaths by police.What changes is which number carries greater significance, of proportionality versus total sums, and this comes down to interpretation.The normative dimension of this interpretation comes down to which interpretation of the evidence makes sense to uswhich statistics carry greater emotional resonanceand this means locating the facts within a narrative that the interpreter in some way belongs to.Using facts to advocate political change is meaningless if they cannot be located within a narrative in which these facts make sense.To put this another way, the veracity of facts within the sciences cannot be divorced from the culture of knowledge within which the sciences operate.This conclusion does not lead to the opposite, that there are no facts outside of this world.Rather, it points to the limits of how far we are able to deploy what count as facts, and to the very real significance of how our narratives help us to interpret the social reality that we are faced with.There is still the possibility to deploy empirical evidence in pursuit of normative goals.What matters is caution in how evidence is used and the acknowledgement that it simply does not matter how many fallible studies one conducts if the research cannot be located within a narrative whereby it makes sense.
The importance of narratives is not just a debate over methodology.Hannah Arendt (1970, 6-7) remarked in her short book, On Violence, about the problematic nature of treating facts of the social and political worlds as if they are the equivalent of the same type of knowledge produced in the natural sciences.The more we pretend that what is done in physics, for example, makes sense in fields outside of the natural sciences, the more we undermine our potential to develop normative arguments in politics.
number of an African Americans being killed by the police.Blue Lives Matter was a counter movement in the United States that focuses on the lethal dangers faced by the police.
Narratives are not everything, but they are important.Evidence matters, but narratives are a form of evidence.It is not as though facts do not exist, or that social scientists should dismiss the scientific method.Rather, the point is that an approach that emphasizes facts unproblematically cannot account for the other fact which is that people interpret the world in a variety of different ways, and that an appeal to factual, ostensibly objective, knowledge is ultimately meaningless when the knowledge fails to make sense according to how we understand ourselves in the world.To put it more pragmatically, it does not matter whether there is or is not an objective real world of facts out there if we do not accept the knowledge that describes this world as valid.
The more we constantly try to get our data right at the expense of all other forms of knowledge production, the less we are able to question those conditions under which we assume that such data can tell us what we want it to.Thus, instead of 'thinking' (Arendt 1971), we treat people and problems like puzzles where the boundaries are clear and easy to find, and all we need to do is put the rest together.With such a dismal view of the human condition, is it any wonder that people have turned away from facts and embraced truthiness?
The current political significance of alternative facts and fake news is in how they undermine the ability to deploy progressive political agendas.It is a mistake to counter them by trying to argue for solely for increased scienticism.Doing so would be self-defeating as it can never escape from the political consequences of falsification and its ability to undermine itself.
Rather, we need to accept the inherent contingency of the world, and recognize the role that narratives play in how we interpret and engage with this world (Baron 2017).All evidence is political, empirical facts and narratives included, and this matters in the study of security if only because the stakes -people's livesin matters of security are quite significant.The next The rise of alternative facts (and that corollary accusation of 'fake news') is closely aligned with the political success of Donald Trump.The centrality of this epistemological position to the Tump administration was made plain when Trump spokesperson Kellyanne Conway went on NBC's Meet the Press on January 22, 2017 and introduced into the national and international lexicon the term 'alternative fact'.The introduction of this nomenclature was directed at re-defining what constitutes a fact as a means to dismiss some factual claims over others, but the political implication of it was that any ostensibly factual claims made by the White House are true regardless of evidence to the contrary.In response, Chuck Todd, the NBC reporter who was speaking with Conway, replied to her statement by saying, 'alternative facts are not facts, they're falsehoods.'Inthis exchange, we are provided with what appears to be the underlying character of the post-truth phenomenon: that empirical claims need not rely on any empirical evidence at all.