6 Comments

The increased availability of books as a mechanism for enculturation is an interesting one. But does the contents of the books matter?

A person's prior history of studying books is a signal of their ability to work at a difficult task without a lot of supervision. Their willingness to spend many hours reading is a signal of their industrious. The content of books will be driven by the fashion and culture of the day, i.e., customer demand is what sells.

Technical books came later because it took a while for nobles to realise that they could gain status by sponsoring their publication, much like they sponsored works of art.

The analysis in 'Setup' pulls equations out of thin air (references would be good), and makes lots of assumptions the free flow of people, ideas and book contents being uniformly of practical use. There does not appear to be much connection with the earlier text.

Expand full comment
author

Thanks for the comments. I think a lot of the model is fairly standard in terms of consumers and firms. That may be why we don't make a ton of references.

About studying as signalling: interesting and simple idea! The fact that the content does actually seem to be about encouraging industriousness, savings, and hard work, is more compatible with our thesis, though. Also, there is a lot of private reading, which doesn't work from a signalling point of view.

Expand full comment

The fact that somebody would spend hours reading in private is the signal. It signals that they are capable and willing of working alone, something an employer seeks in an employee.

The signal is the fact that a person who spent time reading books could discuss them, while a person who had not read them could not do this.

Is this signalling, or a filter for the kind of people who enjoy spending time reading books? Are the kind of people who enjoy reading books the kind of people employers prefer to employ (not if the employer is the army)?

If the content of books had an impact on behavior, then we would expect a range of behaviors to change, based on book contents. All we need is a dataset of the distribution of book contents and a time series of socioeconomic behavior over the corresponding years (perhaps this actually exists, I have no idea).

Expand full comment
author

My point was that signals are more effective if they're observable, which private reading need not be. But yup, we need data. Dittmar (2011) is relevant. Modern experiments on culture and preference change can tell us something too.

Expand full comment

Signals are only effective if they are observable.

If I expound on the works of Nostradamus, then presumably I have read them (I once wrote an App that found his best matching prediction against a sentence given by the user).

With regard to data. Would white-collar crime decrease with an increase in the availability of books (after a suitable lag)?

Expand full comment
author

Maybe, but C17 data on crime (of any kind) is going to be hard. And it has to be disaggregated by place - ideally in Germany, where there's enough variation in the location of printing presses. A single time series can't rule out alternative explanations. There are some suggestive facts, like declines in illegitimacy etc. in some Protestant cities.

Expand full comment