When I joined political science 40 years ago, it was largely non-quantitative; now it has become more pseudo-quantitative, depending on statistical approaches almost alone.
Just try, retroactively, to develop physics, using the prevalent social science methods of today! Newton’s and Maxwell’s laws could not possibly emerge.
Science walks on two legs: asking how things are (including statistical data shuffling), and asking how things should be on logical grounds (using thinking, much helped by graphing).
Social sciences will remain in a relative morass until they develop this second leg, logical models able to make quantitative predictions, and then establish connections among such connections.
How did your interest in Electoral System develop?
As a refugee from Soviet Russian occupation of Estonia, I wanted to understand more about what had happened to my family and country.
So, after a PhD in physics, I took evening courses in political science. I tumbled on the so-called cube law of elections and quickly found the logical reason behind it.
Shifting full time to political science, I have applied the broad methods acquired in physics to many issues, but electoral systems have been among the most fruitful.
This involved a measure to express the effective number of parties, then deducing this number from the number of seats available, first for seats and then for votes, and proceeding to predict the duration of cabinets.
How did you get started in your career?
“With a mere MA in international relations, no second or third rate political science department would hire you. Your only chance is at a top university — they are willing to take chances, if the person looks interesting.”
So I was frankly told at a university that would not consider me. So I forced myself to overcome my natural modesty and sent out 120 letters: “if you think political science is in good shape, discard this letter. But if you think political science, as science, is in a dismal shape, then I am the person to change it.”
Only a school I could not locate on the map responded : “You are a funny kind of a social scientist. We are a funny kind of a School of Social Sciences. Maybe we would be good for each other.”
I have been at the University of California, Irvine ever since, seeing it grow from 5,000 to 35,000 students.
What would the layman find most interesting and accessible within your research?
Start with the simplest building block on which much of my electoral research is based.
Suppose seats in a 25-seat district are allocated by proportional representation, so that even 4% of votes would earn a seat. What is the most likely number of parties to win at least one seat in this district? Think by extremes.
The number of seat-wining parties is at least one and at most 25. The balanced logical guess is 5 parties, with an average of 5 seats per party. And this is the average outcome indeed, worldwide.
Piling up such simple thought experiments, I eventually can predict the average duration of cabinets in a country, simply from the number of seats in the assembly and in the average electoral district.
With all the funding in the world, what would you have researched and how?
I would have been so busy managing the funds and personnel that I would have had no time to think deeply about substantive issues. So I might have produced more pages but fewer interconnected models.
But right now I would look for coauthors for a book on “Country sizes and world population over 5,000 years”, for reworking a textbook manuscript on “Logical Models and Basic Numeracy in Social Sciences” I have been using for 9 years, and for a study of the relationship between per capita GDP and the state share of GDP.