David Zyngier and the Perihelion of Mercury

I have finally given up on Dr David Zyngier on Twitter. He has a tendency to want to disagree with what I write, but his disagreements rarely rise above level DH1 on Graham’s hierarchy. Most recently, after my post about phonics reporting in The Conversation, his response was to question my expertise:

As Paul Graham explains:

“Saying that an author lacks the authority to write about a topic is a variant of ad hominem—and a particularly useless sort, because good ideas often come from outsiders. The question is whether the author is correct or not. If his lack of authority caused him to make mistakes, point those out. And if it didn’t, it’s not a problem.“

However, as part of this conversation, in response to Sara Peden, Zyngier made something approaching a substantive point. This is worth addressing because it highlights what may be a common misconception held by social scientists about the physical sciences:

The physical sciences make use of a series of models, just like the social sciences, and this can send us down a rabbit hole at times.

For instance, a perfectly serviceable model of the mind sees it as consisting of a working memory plus a long-term memory. This model is able to make predictions about learning and is good as far as those predictions go. However, if you make use of this model, you are bound to encounter those who point out that it is a simplification, which it most certainly is. They may talk to you about sensory memory or the structure of working memory. The key point is this: Does adding these elements to the model change the predictions that are being made in any meaningful way? If they do then they are relevant. If they do not then who cares?

This is because nature is highly complex. All scientific models are simplifications. Wisdom is not achieved by drawing on the most complicated models possible, it is achieved by drawing on the simplest models possible.

How is this relevant to Newton’s laws of gravity? Well, they are not ‘scientific fact’. Newton’s laws are a model that work very well under most circumstances. However, they do not predict the perihelion of Mercury. The perihelion is the closest point to the Sun in Mercury’s orbit and it doesn’t stay in one place – it moves around over time.

Newton’s laws do not predict the extent of this movement. However, Einstein’s general theory of relativity does predict the motion of the perihelion of Mercury and this was seen as a great early success for Einstein’s theory.

Does this mean we cast Newton’s laws aside? No. They remain an excellent approximation of general relativity in most circumstances and they have the advantage of being far simpler to apply.

What has this to do with the simple view of reading? The simple view proposes that reading comprehension is the product of decoding ability and verbal comprehension. As a model, it is valid to the extent that it makes accurate predictions. Adding bells and whistles to it to formulate a ‘complex view of reading’ is not a sign of sophistication, it needs to be justified. Do these extra bells and whistles change the predictions of the model? If they do, let’s test those predictions. If they do not then who cares?

Science is the process of modelling reality and testing those models. In this respect, there is no difference between physical and social science.