It's the accepted view. When they was very little artificial light you were dictated by the hours of sunlight available to you. Until the modern era people "on most evening experienced two major intervals of sleep bridged by up yo an hour or more of wakefulness", getting up and doing an hour or twos work, socialising, having sex, drinking, etc, etc. There are diary extracts, contemporary novels talk about them, and wood carvings about them. Willaim Harrison, in his mid-16^th century Description of England referred to "the dull or dead of night, which is midnight, when men be in their first or dead sleep."

Not only was sleep segmented, "napping during the day appears to have been common, with sleep less confined to nocturnal hours than it is in Western Societies today".

When modern lighting was invented the enormous physiological impact of it - and it's absence - was widely agreed upon by contemporary scientists. Charles A. Czeisler, chronobiologist, commented "we are inadvertently taking a drug that affects how we sleep".

In regards to the the theory that the industrial revolution put an end to segmented sleep, we can see that artificial lighting was indeed to blame because of how it decreased in frequency among the upper classes first as "there was a shortage of artificial lighting... [that] feel hardest on the lower and middle classes." Tallow candles were an expense that many could not afford too regularly. This class divide becomes increasingly evident by the late 17th century "when both artificial lighting and the voge late hour grew more prevalent among affluent households" and saw the death of their second sleep.

A.R. Ekirch wrote an article in the American Historical Review April 2001, called "Sleep we have lost: Pre-Industrial Slumber in the British Isles" and it is a great read for this topic.

Follow-up: if segmented sleep is accepted, has their been any studies or theories about how it differed between people in different latitudes? My pre-industrial ancestors here in Finland had much different natural light patterns than someone in North Africa, for example.

You might be interested in a previous answer I wrote about polyphasic sleep among modern Amazonian foragers.

Unfortunately, there has not been a tremendous volume of work conducted on the anthropology/ecology of sleep. Carol Worthman's work might be your best bet for more information, but I don't know if she has published much recently. Also, check out this Discover article for a brief overview of cultural variation in sleep patterns, specifically modern foragers.

You should ask this on /r/AskScience. There is not enough historical evidence one way or the other, except the work of Ekirch who is quoted in every single one of the articles about segmented sleep. But if you read the literature, there are dozens of articles that flatly contradict Ekirch, including scientifically done empirical sleep studies as well as a lot of anthropological research.

My impression is that while segmented sleep may have existed in certain specific conditions (winter in high latitudes, where nights were very long), these represent a tiny fraction of the human population. On the vast majority of the globe, such conditions didn't exist, and probably neither did segmented sleep. There is a logic to breaking up your sleep into two phases if the night is sixteen hours long. But switch the seasons to make it summer, and the days are sixteen hours long instead. The same logic now dictates that you might want to use your few hours of darkness to get some uninterrupted sleep rather than being awake in the dark and sleeping when it's light.

How much evidence of this sleep schedule amongst present day human societies that do not use modern technology?