Skip to content

Prof. Elizabeth Anderson’s Uehiro Lectures: Lecture 2 Summary – “Improving Political Discourse (1): Re-learning how to talk about facts across group identities”

Prof. Elizabeth Anderson’s second Uehiro lecture focuses on how we can overcome obstacles to fact-based political discourse. In particular, the lecture concerns how we might prevent identity-expressive discourse (a term introduced in the first lecture; see summary of lecture 1 below) from displacing the discussion of facts and evidence in public discourse, and how we might overcome the shameless lying and disinformation campaigns of populist populations. Over the course of the lecture, Anderson illustrates her analysis with illuminating cases studies, and finishes by providing her own solutions to the problem at hand, drawing on Cultural Cognition theory, John Dewey’s cultural conception of democracy, and emerging data from deliberative polling studies. You can find a recording of the lecture here

Anderson begins by suggesting that the problem under discussion can be recast into the following two questions:

  1. How can we learn to set aside identity expressive discourse so as to orient political discussion to factual discussion?
  2. How can we orient factual discourse in an evidence directed way?

To get a handle on these questions, she highlights political discourse about climate change as a case study of the issue under discussion. Despite evidence of widespread agreement amongst republicans and democrats in the 1990s, data suggests that the two parties have become increasingly split over time with respect to their views about the risks of climate change. Why is this the case?

For Anderson, an answer is suggested by Dan Kahan’s work on Cultural Cognition. One of the core themes of Cultural Cognition is that cultural values shape how individuals perceive risk. We are biased towards assessing risk in ways that affirm our cultural identities, and our views of how society ought to be organised. On this view then, statements about risk are primarily declarations of identity, and do not necessarily convey beliefs. As Anderson notes, this means that statements of risk can be examples of identity expressive discourse.

More specifically, some work in Cultural Cognition (by Mary Douglas and Aaron Wildavsky) suggests that views about risks appear to reliably track certain cultural views. This was illustrated with a diagram that I shall try and roughly explain: The diagram included a vertical axis bisecting a horizontal axis, thus illustrating four quadrants. The left of the horizontal axis was labelled ‘individualism’, whilst the right was labelled ‘communitarianism’. The top of the vertical axis was labelled ‘hierarchy’, whilst the bottom was labelled ‘egalitarianism’. As such, any given individual could be placed in one of the four quadrants in accordance with where they fall on the scale of these two axes of competing cultural views. For instance, someone could score highly for hierarchy and individualism, and would thus be placed in the top left quadrant – we might refer to them as a ‘hierarchical individualist’.

The key finding from this work is that an individual’s placement on this cultural grid seems to reliably track their views about risks. For instance, the data suggests that hierarchical individualists believe that climate change and nuclear power pose low levels of risk, whilst gun control laws and immigration pose high levels of risk. Interestingly, Egalitarian-Communitarians (who are in the bottom right quadrant, and represent individuals with the opposite views to hierarchical individualists on both axes) hold the exact opposite views on all of these risks.

Anderson relates these findings to a causal model of how cultural world-views can affect our view of certain activities and their risks. On the model she presents, cultural world-views can give rise to emotional responses; but these responses are also modulated by the manner in which different activities are represented to the individual. Crucially, these emotional responses influence the individual to develop an affect-driven cost-benefit assessment of the activity. Aware that the framing of different activities can compound this affective influence, Anderson notes the possibility that some parties may intentionally seek to manipulate this by propagating antagonistic memes that encourage people with different world-views to distrust each other through, encouraging conflict and triggering further affective influence. She terms such parties ‘conflict entrepreneurs’.

For Anderson, this model offers a causal explanation of why we are currently so conflicted. She also identifies populist politicians as prominent conflict entrepreneurs, but also recognises that those with competing cultural views often add fuel to the fire of political polarisation, by themselves engaging in antagonistic memes, and joining the populist fight in these terms. For an amusing (but also alarming) illustration of how this played out in the escalating ‘bumper sticker wars’ in the US, I recommend listening to the recording when it becomes available!

Given this picture of the causes of our current polarisation, what hope is there? Anderson suggests that it can be found in John Dewey’s cultural conception of democracy. Going beyond purely proceduralist accounts of democracy, which reduce democracy to the constitutional structures that enable democratic decision-making, Dewey’s understanding calls for a broader understanding of what democracy involves. In addition to the aforementioned constitutional structures and procedures, democracy in Dewey’s view requires a democratic culture, one that involves the orientation of citizens to one and other in a co-operative discourse about the issues of the day (not necessarily just in the political sphere). A core feature of the view is that cultural democracy requires us to learn how to treat others as equals in the public sphere, and to treat those with whom we disagree as those from whom we might learn.

Anderson is optimistic that we are capable of enacting this sort of cultural democracy, adverting to recent data from lab-based deliberative polling. In the “America in one room” study, 526 random US voters engaged in 4 days of collective discussion of polarising issues (including healthcare, immigration etc). Each participant was provided with a 55-page briefing book beforehand, developed by experts from both the Republican and Democrat parties (there was, however, no identifying info on the briefing book, on specific ideas, or on individual participants). After 4 days, the data suggested that considerably more participants agreed with the statement that “US democracy is working well” after the deliberative polling exercise. Furthermore, the vast majority agreed with the statement that they “learned a lot about people very different from me”. Perhaps most strikingly though, the data also suggested that Republican attitudes towards the various contentious issues had dramatically softened.

What explains this change? Returning to the causal model described above, Anderson suggests that deliberative polling essentially removes the influence of conflict entrepreneurs from the discussion, and replaces it with the personal stories of diverse citizens. Crucially, this serves to activate a set of empathetic emotional responses that are markedly different to those triggered by antagonistic memes.

Anderson goes on to suggest that further support for her optimism about cultural democracy can be found in the Climate Change Adaption project in Florida. This was a real world case of non-political groups mobilising to organise public meetings to discuss solutions to the shared problem of climate change, and Florida’s particular vulnerability to rising sea levels. Crucially, organisers of the discussion refused to frame the issue in partisan terms, and fought back against antagonistic discussants who sought to reframe the debate in this manner. Notably, this project led to the development of a bipartisan agreement on a plan to deal with the shared problem of climate change.

Such co-operative debate leads us away from a discussion based on identity-expressive attitudes because it diminishes the influence of the individual’s world-view, and replaces it with the expressed shared concern of diverse fellow citizens. As such, on Anderson’s causal model, the affect based response will not triggered in this kind of environment. The approach thus serves to disentangle the discussion of what people know, from who they are. It allows us to resist the influence of cultural cognition in our political discourse,

Finally, Anderson suggests that a further way to encourage better political discourse is to essentially harness the power of cultural cognition, and find solutions to shared problems that affirm cultural identities. For instance, one could attempt to bring hierarchical individualists on board with attempts to reduce their carbon footprint by encouraging their self-sufficiency, a value that is central to their cultural identity, but which is also likely to be compatible with environmentalist interventions

In conclusion to this discussion of how to overcome obstacles to fact-based political discussion, Anderson offers the following 5 concrete practical recommendations:

  1. Promote face-to-face discussion of political issues by diverse stakeholders.
  2. Do so in contexts that avoid or marginalise conflict entrepreneurs and antagonistic memes.
  3. Encourage people to speak from personal experience about personal impacts – Such discussion should avoid abstract moral principles and science, and instead focus on promoting empathetic responses.
  4. Identify shared concerns that require shared solutions
  5. Look for creative revision of understanding both problems and solutions in ways that avoid antagonistic framing, and accommodate and affirm diverse identities.

In this far-reaching lecture, Prof. Anderson clearly illuminated every aspect of one of the key problems facing our political discourse, and provided some compelling practical solutions. However, I wonder if the particular features of the Florida example raise some concerns about the large-scale implementation of recommendation 4 above, and for the prospects of cultural democracy more generally. In some cases, our shared concerns affect us each to a more or less equal extent, and we may each have more or less equal responsibility for the issue of concern. In such cases, it may be quite plausible and natural for communities to cooperatively engage in cultural democratic projects and to develop shared solutions to the shared problem. However, is there also scope for optimism about cultural democracy with regards to global problems that are in some sense shared, but that are perhaps likely to affect some far more than others (at least in the short term), and for which different cultures and communities may be perceived to have quite different degrees of responsibility?

Share on