One problem with making individual choice the foundation of society is there is increasing recognition that people are not necessarily very good at it - certainly for more complex choices. It is easy to choose which is the cheaper gallon of milk at the supermarket. It is much more difficult to figure out your 401k plan or give up smoking.
One of the most prominent of these new approaches is behavioral economics, which says we have systemic biases and errors in our thinking.
A related - and older - argument, which seems very relevant these days, is the existence of bubbles in markets. We are suffering the aftermath of a vast housing bubble, when millions of people lost all sense of proportion on housing prices.
The history of bubbles goes back hundreds of years - for example, see the classic Extraordinary Popular Delusions and The Madness of Crowds. One of the best works in economics on the subject is Charles Kindleburger's Manias, Panics, and Crashes: A History of Financial Crises (Wiley Investment Classics).
Far from being rational choosers, we are often swept up in the mood of the moment.
It is important to say this does not justify authoritarian choices by experts on other people's behalf. The experts are generally little better at making choices. But it does mean we have to work with what we have got, what Kant called "the crooked timber of humanity", rather than imagine people are better at making choices than they actually are.