Jump to comment:

• Model Hazard?
  Temitope Fisayo, Alexander Gray
  Published on: 3 May 2020

• Published on: 3 May 2020

  Model Hazard?

  • Temitope Fisayo, Medical Student King's College London
  • Other Contributors:
    • Alexander Gray, Economist

  Richardson (1) argues three substantive points:
  1. Models are merely fables dressed in formal language.
  2. Fables are unscientific.
  3. Models serve as epistemic confines to our understanding.

  We argue that 2., a premise he makes implicitly, is wrong. Formal language in fables cannot produce an ‘illusion’ of scientific-ness, because there is no division between ‘fables’ and ‘science’. We suggest that scientific models are stories (2) in some real sense, and therefore it does not make sense to say that models are unscientific because they are fables. Science is composed of a complex web of interacting models (stories) whose aims are to explain and understand the world. This would be consistent with Sugden’s (3) view of economic models as credible worlds.
  Richardson cites Rubinstein (4) to buttress his argument that models are merely fictions. This misuses Rubinstein, who argues ‘The models presented… are nothing but fables. Neither of them describes reality, but both of them still describe something from reality… studying both of them together helps to some extent in understanding economic mechanisms.’ (p.182). It does not seem fair to brand models as ‘merely’ fables on this reading, and nor does this give us licence to dismiss fables as unscientific.
  Epidemiologists often make significant assumptions in order to model disease progression. Many parameters are unknown, and there are often practical constraints to modelling significant hete...

  Show More

  Richardson (1) argues three substantive points:
  1. Models are merely fables dressed in formal language.
  2. Fables are unscientific.
  3. Models serve as epistemic confines to our understanding.

  We argue that 2., a premise he makes implicitly, is wrong. Formal language in fables cannot produce an ‘illusion’ of scientific-ness, because there is no division between ‘fables’ and ‘science’. We suggest that scientific models are stories (2) in some real sense, and therefore it does not make sense to say that models are unscientific because they are fables. Science is composed of a complex web of interacting models (stories) whose aims are to explain and understand the world. This would be consistent with Sugden’s (3) view of economic models as credible worlds.
  Richardson cites Rubinstein (4) to buttress his argument that models are merely fictions. This misuses Rubinstein, who argues ‘The models presented… are nothing but fables. Neither of them describes reality, but both of them still describe something from reality… studying both of them together helps to some extent in understanding economic mechanisms.’ (p.182). It does not seem fair to brand models as ‘merely’ fables on this reading, and nor does this give us licence to dismiss fables as unscientific.
  Epidemiologists often make significant assumptions in order to model disease progression. Many parameters are unknown, and there are often practical constraints to modelling significant heterogeneity in the population, for example computing power. When an epidemiologist assumes ‘symptomatic individuals are 50% more infectious than asymptomatic individuals,’ [https://www.imperial.ac.uk/media/imperial-college/medicine/mrc-gida/2020... they are explaining how the world might be under this assumption. This is just one example, but we are not the first to suggest that models in science describe the world as it might be rather than necessarily as it is (5–7).
  Therefore, science is composed of a network of interconnected stories (models), and it does not make sense to think that because models are stories, they are unscientific.
  Richardson’s description of models appears to grant them agency (8). This means it is the models themselves that warp our understanding of the spread of SARS-CoV-2. He thinks this happens in a similar fashion to philanthropists obscuring economic exploitation. We argue, however, that this is an unhelpful inflation of models’ agency: if models restrict COVID-19 discourse and impose epistemic confines, it is humans, political actors, who make them do so. Thus, just as billionaire philanthropists (humans) marginalise discussions over more equitable taxation regimes, it is human political actors who instrumentalise models to suppress contemplation of potential worlds. Concern over one or another model’s agency seems to lead only to more discussion over models - if the model has too much or too little agency, this is a problem for the model, and the question of how humans use models is vanished.
  Even if we accept that models are indeed agentic, we argue his remedy of ‘liberation by model’ is misplaced. More modelling with ‘radical wealth redistribution as its moral’ feels unlikely to move discussion away from the modelling and towards the other causes of poor health. Instead, we suggest looking to other materials to aid our understanding of COVID-19. Models must sit alongside (e.g.) history and politics as tools to usefully describe what is happening (and what could) - but models cannot tell us what ought to.

  References

  1. Richardson ET. Pandemicity, COVID-19 and the limits of public health ‘science.’ BMJ Glob Heal [Internet]. 2020 Apr 1;5(4):e002571. Available from: http://gh.bmj.com/content/5/4/e002571.abstract
  2. Frigg R. Models and Fiction. Synthese [Internet]. 2010 Apr 18;172(2):251–68. Available from: http://www.jstor.org/stable/40496038
  3. Sugden R. Credible worlds: the status of theoretical models in economics. J Econ Methodol [Internet]. 2000 Jan 1;7(1):1–31. Available from: https://doi.org/10.1080/135017800362220
  4. Rubinstein A. Economic fables. Open book publishers; 2012.
  5. Frigg R, Nguyen J. The turn of the valve: representing with material models. Eur J Philos Sci [Internet]. 2018;8(2):205–24. Available from: https://doi.org/10.1007/s13194-017-0182-4
  6. Frigg R, Hartmann S. Models in Science. In: The Stanford Encyclopedia of Philosophy [Internet]. Spring 202. Metaphysics Research Lab, Stanford University; 2020. Available from: https://plato.stanford.edu/archives/spr2020/entries/models-science/
  7. Rhodes T, Lancaster K, Rosengarten M. A model society: maths, models and expertise in viral outbreaks. Crit Public Health [Internet]. 2020 Mar 31;1–4. Available from: https://doi.org/10.1080/09581596.2020.1748310
  8. Latour B. On actor-network theory: A few clarifications. Soz Welt [Internet]. 1996 Apr 18;47(4):369–81. Available from: http://www.jstor.org/stable/40878163

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top