Did humans live more densely in the past, and if so, how much denser?

Question

I know family sizes were larger in the past, "average U.S. household – from 5.79 people per household in 1790 to 2.58 in 2010." but how closely did those family members live in their houses?

I'd settle for an average, or failing that, a well-sourced densest family disregarding any concerns about how richer families can support more people, etc.

And I'm looking for values in "persons / 2-dimensional area", where the "2-dimensional area" is a house, or equivalent structure, a.k.a a dwelling.

Lastly, I'm looking for the densest family of the past, a.k.a, I don't care where or when it is, I just want the most family members packed into the smallest dwelling of their own volition. Slave-owning families, in the Americas or Greece or anywhere else don't count, unless you exclude the slaves from the family member count.

Answer 1

(Note this answer cannot answer the 'maximum people per square foot/meter of living space', but presents a more general bit of useful information concerning the variation of population typically found within a historical household or family.)

Historical demographics rely on various means to be able to estimate historical populations within cites. Direct census type information is rarely available, so individuals trying to calculate a cities population at a given point in history often need to collect examples of information they can find, and then create reasonable formulas to enable population estimates to be made when they find such data, or to enable validation of other historical data encountered.

One epic study (representing 30 years worth of research) can be found in Chandler and Fox, 3000 years of urban growth. (If you have an interest in medieval demographics, this should be your bible.) On page 6 in the opening chapter is a table which represents the relevant types of data this study uncovered on the typical number of inhabitants at various locations and times:

You can see the various data that was collected was from different sources which included counts of households, houses, taxpayer or families. Every source for each cities' information is detailed in the book. So in comparison to your opening statistics of 2.58 to 5.9 people per household, actual data on late medieval to early modern era populations typically shows values above that. Chandler states that Genoa was the densest city in Europe, and he shows a population of 15 people per house in 1460, well above the 5.79 from the opening numbers.

Chandlers population density for Genoa is listed as over 600 people per hectare (abt. 243 people per acre.) We can compare this with modern density figures for Manhattan of 74,780.7/sq mi (28,873.0/km2) which equals roughly 116 people per acre (288 per ha). So Genoa in 1460 was more than twice as densely populated as Manhattan is today.

So, did people in the past 'live more densely' than today? It varied by circumstance, but there were definitely locations where the population was much more dense than today.

(Note Chandler and Fox must be borrowed with login to access the full text.)

Answer 2

Inuit igloos historically housed entire families and rarely exceeded 4.5 m in diameter, an area of 16 m^2 or about 160 sq. ft. With family sizes often above a dozen, that's in excess of 70 people per 1000 sq. ft. I doubt you will exceed that in non-Arctic conditions. Even with all that body heat, and a small oil lamp for additional heat, temperatures inside an igloo might be barely above freezing with outside temperatures below -40 C.

In contrast, Charles Garland and Herbert S. Klein in The Allotment of Space for Slaves aboard Eighteenth-Century British Slave Ships report that the UK in 1799 Parliament prescribed a minimum of 8 sq. ft. per slave, or 125 slaves per 1000 sq. ft., while other calculations suggest that in practice only 5 or 6 sq. ft. was allocated per slave, a rate of 165 to 200 per 1000 sq. ft.