Deep Thinking, Reading, and Distractions
In an article provocatively titled ”Is Google Making Us Stupid?” Nicholas Carr discusses the influence of online reading and search on the way we think. It is a well written article really worth reading, but while I agree with many of the arguments, there are some points that I disagree with.
The observation on which the discussion is founded is that reading on the Internet changes the way how we read an think. The reading process is now constantly interrupted, as the mind drifts off, “looking for something else to do”. Anecdotes and research suggest that many people experience this change. A study cited in the article, found that users of online resources of research literature were reading with less depth, “hopping from one source to another.” In a later post to his blog, Carr points to further research that confirms this trend. In that post, he points out one likely cause for this shallowing research behavior: Search engines are popularity engines that concentrate attention rather than expanding it. The recently launched search engine Cuil claims to use different metrics, but is yet unclear how much impact Cuil will have.
The connection between reading habits and thinking is based on the hypothesis that media are not only technical devices for transporting information, but also “shape the process of thought”. This is an old argument that has been discussed in many iterations since the invention of writing (e.g. for the invention of the printing press). The effect of the Internet, as Carr perceives it, is to reduce the capacity for concentration and contemplation.
Reading on the Internet may even cause a different wiring of the brain than reading books does. While this seems to be a surprising suggestion, neurological research has found that what you do and learn indeed shapes the brain physically.
After this exposition Carr explores the origins of this development, which are in his opinion (1) the introduction of Taylorism into “knowledge work”, and (2) the quest by programmers and especially Google to develop the Internet into an artificial intelligence. While (1) is an interesting point well worth exploring and evaluating its merit, (2) is seems like far-fetched fear-mongering to me.
The ethics of Taylorism are controversial in any field, but in manufacturing it has formed the basis for increased productivity. In “knowledge work”, it seems much less clear if a defined process leads to better results at all. The problems in this domain start with measuring the performance, a corner stone of Taylor’s system.
Mostly Harmless
I call Carr’s warnings about emerging artificial intelligence fear-mongering, as AI in my eyes produces either boastful claims about lofty goals or rather mundane things such as Lisp or pattern matching that, useful though they may be, fall miles short of the claims.
Who is “The Net”?
Another thing bothering me about the article is that Carr uses the term “the Net” almost as if it were a person with an agenda of its own. “When the Net absorbs a medium … it injects the medium’s content with hyperlinks, blinking ads, and other digital gewgaw …”, sounds a lot more dire than: “When a medium is adapted to the Internet, content providers inject the content with hyperlinks, blinking ads, and other digital gewgaw.” It obscures the economical and social forces behind the development.
Confusion about active forces also weakens the discussion about interruptions that lead to shallow reading: “A new e-mail message, for instance, may announce its arrival … The result is to scatter our attention and diffuse our concentration.” Why does the e-mail announce it’s presence? If I wanted to be pedantic here, I would say that your mail user agent announces the presence, not the e-mail itself. But even when we leave such nitpicking aside, the announcement is only there because the software is configured to notify the user. There is nothing else that forces the program to show an announcement; e-mail is a much more asynchronous way of communication than e.g. the older medium of telephony.
So instead of lamenting that technology forces us to do things, we should ask: “Why do we use the technology, even when we know how to turn it off?”
What to do about it?
This leads me to an issue that is really missing from the article. The objections discussed above are actually more about the tone of the text and some details, I do not disagree that there is a change in how we read that changes how we think. But what are the consequences we should draw from this realization?
One thing we must ask ourselves is, why don’t we actively turn off all the announcements and attention-grabbers at least part of the time? For parts of our lives, each of us can decide this individually. In “Disconnecting Distractions” Paul Graham writes that for working he uses a computer that has no Internet connection to keep himself from procrastinating by using the web. In other parts of our lives, external forces govern (at least to some degree) how we have to configure the potential for distractions. In “Brain Rules” John Medina discusses the inefficiency of workplaces full of distractions and suggests alternatives. This problem is only to some extend related to the Internet; talking with colleagues can have the same effect.
These are, however, only superficial measures and a natural inclination for laziness and procrastination is working against it. Maybe we could design technology in such a way that it focuses attention instead of scattering it. This is possibly not the kind of thing Nicholas Carr would suggest, but to me it seems to be a worthwhile approach. After all, we make technology to help us, and if it doesn’t work as we intended, we have to make an effort to fix it. More information does not have to be a bad thing for attention; as Edward Tufte states: “To clarify, add detail”.
This is not something to solve once and for all with a single technology, of course. It is rather something to be considered whenever designing a user interface – like security that has to be considered whenever designing a system.
