Thursday, September 20, 2007

Week 6: Cultural, linguistic, & genetic diversity patterns 1

For the coming week, please read the following, ideally in this order:

1. Moore et al. 2002. The distribution of biological & cultural diversity in Africa

2. Collard & Foley. 2002. Latitudinal patterns & environmental determinants of recent human cultural diversity

3. Pagel & Mace. 2004. The cultural wealth of nations.

4. (Optional) Nettle. 1998. Explaining global patterns of language diversity. [Nettle

5. (Optional) Cashdan. 2001. Ethnic diversity and its environmental determinants.

6. (Very Optional) Serre & Paabo. 2004. Evidence for gradients of human genetic diversity

Maps: (top) topography; (2nd) annual precipitation; (3rd) vascular plant species diversity, (bottom) world language diversity

Click on these maps to see them in greater detail.


Humans are part of ecological systems and have various ecological relationships and associated life history patterns. Then again, unlike other large mammals, humans have permanent settlements on every continent and in almost every known terrestrial environment. Humans are relatively genetically homogenous, yet we display astonishing cultural and linguistic variability that follows clear geographic patterns. Is that variability a function of environmental variables, like climate or ecosystem type? Where human cultural patterns mirror biogeographic patterns of other species, are humans responding to the same forces? In the same ways? Why or why not, and how do you test such questions anyway? To start with, how do you fruitfully think about variation in cultural and social patterns—as discrete entities, as clines, or otherwise? Are these differences comparable to other biogeographic differences, and why? More broadly, what is the relationship between culture, language, and environment? Finally, how does human life history, such as lifespan and learning period, factor into such patterns and relationships?

These are some of the questions we’ll ponder and discuss over the coming two weeks. You will struggle with these ideas, occasionally you will triumph, you will laugh, you will cry. I’ll provide Kleenex.

On a more serious note, carefully read the required articles and, if time, the abstracts and figures in the optional articles. Moore et al., 2002, focuses on the relationship between language “richness” (see Glossary below) and species richness in Africa. Collard & Foley, 2002, takes a macroecological look at latitudinal patterns of cultural diversity and possible environmental determinants. Pagel & Mace, 2004, is a good, readily digestible review of current hypotheses on the relationship between culturo-linguistic diversity and environmental variation.

The optional articles provide good reference for the required ones and offer additional hypotheses. Nettle is very active in debates about linguistic diversity, and this paper gives his most widely-known view, which Pagel & Mace summarize and reference. Cashdan has done lots of work with hunter-gatherers and looks carefully at several ecological variables. If you are interested in reading more or following this topic for your project, Nettle and Cashdan are great starting points. Serre & Paabo provides some good recent context on the relationship between human genetic & linguistic patterns, but it’s useful here just for context on how human mating and migration patterns relate to linguistic patterns; we’ll focus explicitly on this topic next week.

If you come across terms in the papers you don't know, consult the Glossary below for reference and understanding. You don't need to memorize these terms, but understanding these concepts will help with the papers.



p.s. Also consider that languages and cultures, like species, are disappearing at a clip. A recent NYT article (

estimated that one of humanity’s approximately 6900 languages is “lost” every two weeks. Indeed, language and cultural documentation and preservation is an active field, and a major thrust is the tight link between biological and culture-linguistic diversity.


allopatry – separation of a population into two or more groups via a physical barrier, such as a mountain chain or a river; allopatric speciation is the formation of two or more species as such physically divided populations genetically and behaviorally diverge

biome – one of the Earth’s main, usually terrestrial, ecosystem types, such as desert, tropical rain forest, and tall-grass prairie

carrying capacity – the maximum population size of a given species that a given environment can support relatively stably over time

correlation – in statistics, the strength and direction of the linear relationship between two variables

diversity, species—in ecology, a measure of the diversity of species in a given area that considers both “richness,” or number of species, and “evenness,” or evenness of the population sizes of those species

endemic—native and often unique to a given area

evapotranspiration—the transfer of water vapor from Earth's surface to the air from evaporation and transpiration (transpiration is the release of water vapor from plant leaves & stems as a byproduct of photosynthesis) ; potential evapotranspiration is the maximum water vapor an ecosystem could release if it got enough rain.; actual evapotranspiration is the amount of water vapor an ecosystem does release given the average precipitation it usually gets.

genetic marker— a known DNA sequence whose presence and variation can be used to infer the degree of relatedness among groups of organisms

horizontal transmission— transmission, as of a word’s meaning or a custom, within a society other than between parent and child (i.e. other than “down the generations”)

PCA (principal components analysis)— a statistical technique for reducing the number of dimensions of interest among a group of variables; generally, it helps you see which attributes of a dataset contribute most to the patterns of interest; it’s a complex idea that you don’t really need to know…I’m just defining it generally because a paper refers to it

phylogenetic tree—a tree-like diagram showing the evolutionary relationships among species or other entities sharing a common ancestor

population in ecology, the individuals of a given species living together and interbreeding in a given area

productivity—in ecology, the amount of new living (plant) tissue produced, usually over a year’s time[so warm areas with high precipitation produce lots of new plant material per unit time and so have high productivity]; Gross primary productivity is the total CO2 transformed into living tissue, while net primary productivity is this "gross" amount minus the amount used for the plant's metabolism

reproductive isolation—the separation of two or more groups from possible interbreeding; over time, reproductive isolation often promotes the formation of distinct species, or speciation

richness – number or variety, as of species, languages, or cultures. (e.g. The endemic plant species richness of New Mexico is simply the number of plant species native to the state.)

Rapoport’s rule – the tendency for species range size to increase with increasing latitude (i.e. species at high latitudes tend to occupy large ranges and vice versa)

spatial autocorrelation— the degree of spatial, usually geographic, clustering among features of interest; for example, closely related species tend to have ranges that are spatially autocorrelated simply because they descend from a parent species that presumably lived in the general area (i.e. their ranges aren’t clustered simply because the species have similar niches); very generally, it says two points close together are more likely to be similar than two points further apart

species-area relationship – the positive relationship between the size of a habitat and the number of species it can support and that you'll find there [how might this logic be extended to human ecology?]

sympatry – (literally, “same father”) deriving from a single parent species and living in the same area; so sympatric speciation is the formation of two or more species from the same “parent species” without the help of some obvious barrier to interbreeding

vertical transmission—transmission, as of a word’s meaning or a custom, between parent and child (i.e. “down the generations”)


paul said...
This comment has been removed by the author.
paul said...

The density of languages in the Caucasus Mtns and in the Himalayas (Nepal, Tibet) is interesting. It looks like plant diversity it high there too? (hard to tell.)

Steep, inaccessible mountains can make absolute difference translate to a lot greater relative distances. And inaccessible areas are more resistant to invasion. So when the Persians, Greeks, Slavs, Turks, Mongols swept across the Middle East and Central Asia at various points, the old languages spoken up in the Caucasus never got washed away.

Same effect on the differentiation of Tajik dialects in the Pamir mtns. There's alot more diversity than in the Tajik/Turkic speakers living down on the plains.

Are the New Guinea highlands similar? I donno...

dodegard said...


My question was along the same line. Collard and Foley talk about New Guinea having a high rate of cultural differentiation (page 9).

Why would the rates of cultural diversity be denser/higher on islands? I wasn't familiar with that specific islands geography, but that seems like a pretty good reason.

dodegard said...

What are the features of a language?

Wenyun said...

The biodiversity hot spots, nowadays, is also related to remote place. Languates density in those areas may be also contributed by remoteness and isolation which is hard to be disturbed by outside world. Before China was unified by the first king of Qin dynasity, there were futher more languages in that big area. By the way, what is language? Does it have to have writen characters or not?

Fred Whiteman said...

In Moore et al, the authors point out that different species can overlap in ranges, but different languages generally do not. They also point out that history may be a confounding factor, because history may affect species richness differently from language richness. Would a study which estimated language richness and species in an area over time, rather than at a single moment, alleviate these issues?

Deepta said...

In the data section of the Moore article, the authors mention that they leave widespread languages such as Arabic out of their analysis, but I did not see an explanation of why. Wouldn't this alter the results substantially?

Fred Whiteman said...

After reading both Collard and Foley and Moore et al., I’m curious about how the patterns of cultural and species richness would vary when comparing different regions with similar latitudes but different histories. For example, how would Africa compare to South America, where a megafaunal extinction occurred in the Pleistocene? How would Eurasia, which has been populated by humans for much longer, compare to North America?

Justin Smith said...

Continuing dodegard's question--might there be some sort of island rule for culture, a social insular dwarfism? Specifically, are human social units smaller/more dense on islands than on continents such that it takes less area to support a distinct culture on an island?

dtinucci said...

Before we accept climatic/resource variability as the primary cause of language/cultural diversification (contra Nettle)we should consider how geography affects ontogeny. Organisms conform to environments in ways that transform the phenotype. For instance, once perissodactyls invaded the great expanses of high latitudes the norm of reaction tracked the environment by selecting for longer legs. In the same vein, I would argue that the tropical density of biomass precludes contact - and absent easy communications and tranport, along with trade and interaction they facilitate, states integrating numerous small polities will not form. In other words,'what you see is what you get'. Regional command is all about control - which will not materialize if you remain confined to a densely patched habitat.

If time allows, please go over Collard & Foley's phase (p.10) and explain "the human pattern is the microevolutionary, microecological foundations for the macroecological pattern seen in biological communities"?

Steven M. said...

On page 1648 more mentions that only amphibians do not co vary with language diversity. What makes amphibians different? Could understanding why amphibians do not show the same trend with respect to language diversity increase understanding of the environmental variables that affect language diversity?

Senorita Myra said...

Okay, in the Moore et al paper, they talk about null models and hypothesis. i do not understand this. I looked up the definition, but still cannot seem to wrap my head around it. Can you explain how it relates to this specific paper? It seems to be very important what the null model predicts if it can determine alot about the weight other explanations have compared to Moores.

tlvandeest said...

The correlation between species richness and language richness in the Collard and Foley article is related to the rainfall of the area. This is also supported by the wetness correlation found by Moore et al. In terms of the predictive nature of the model created, can these correlations be used to assess the cultural history of an area in terms of estimating the diversity of past human cultures? Would these models hold for past environmental conditions as well as current climates? Can it be said that ancestral hominins were reacting to the same environmental conditions with diversified cultures in areas of high rainfall or wetness?

Dan said...

In the Pagel, Mace paper they state that we share 98% of genes with chimps because our genome is closed off. So why do we like to identify ourselves with such proverbs as "give us your weak, your tired etc..". Are we liars? do we really want who ever will to come? Must those who come assimilate or can they keep all or part of their own culture? Is horizontal transmission of culture really as difficult as they imply? It seems to me that units of culture much like genes will be favored the better they are able to spread.

helen elizabeth said...

In addition to Collard and Foley's mention of New Guinea, Madagascar is an excellent example of an island with high rates of cultural differentiation -- in line with what Paul was saying. Species diversity has been pretty well documented there and their cultural/ linguistic patterns are pretty interesting as well. However, I would suggest checking this out if you want to see some cool models concerning human biogeography and linguistic/cultural patterns: CULTURAL ECOGEOGRAPHY: MODELING INDIGENOUS CULTURAL DIVERSITY PATTERNS, which was presented earlier this year by our own instructors Bill and Jordan.

paul said...

What about dropping the need for discrete units of analysis?

How about, for each grid on the map, having an index of the phenotypic/cultural variance in a trait/behavior that exists. And explain that.

For 35N21W, how many different words are there for sky? If there are a bunch of languages, there will be a bunch of words. If there is a particularly rich language likewise. How many different ways are there to make a living? What is the variance in the length of women's hair? How many different words are spoken in that little grid?

NYC would show up as the most culturally diverse place on earth. (Would it? Hmmm... India and New Guinea are crazy places too). Omaha, not so much.

And sure, the past 500 years have been crazy, lots of movement, not alot of time for patterns to settle out. So maybe freeze the world at 1000. What do you see?

paul said...

Deepta, I was also surprised they left out Arabic, Swahili, etc. And no explanation!

I guess they're dividing up the continent, and if the language ranges across a quarter of the map, where do you put its little flag?

Just a guess. I think you point to an important weakness in the methodology. They need a more sophisticated way of mapping language diversity.

Wenyun said...

How Price did the survey for the Atlas of World Cultures? Compare to other continents Asia seems to have less cultures overall.

Michael said...

Mace and Pagel rely heavily on an article published by Sokal et al. in 1990 to support their argument that language differences may hinder gene flow. A result of my own curiosity, I read through this paper to distinguish what the orange lines in on figure 2 stood for, as these two seemed to have been overlooked. The first, bisecting Iceland, is a result of the country’s history. The western portion had a much higher proportion of settlers sailing from Scandinavia by way of Ireland and Scotland, while the east remained predominately Norse. The second, a line separating Greece into north and south, is a result of sampling. Those sampled were thought to be Greek speakers of Anatolian descent resettled after World War I.

paul said...

I think Pagel and Mace underplay horizontal cultural transmission. We’re only going to have about 5 cultures in a few years, and all 5 are going to look pretty similar, and it ain’t because Westerners are out-reproducing everyone else. When they say that “worries about cultural swamping are overstated”, it makes me wonder how much time they’ve spent out and about.

Dan said...

Paul, I agree
It seems that they are trying so hard to describe cultures as these exclusive groups. That they are wary of outsiders with strong in-group cohesion. They forget that cultures also like to spread. However, they do state that cultural diversity was much greater in the recent past indicating a trend toward homogeneity.

Are there parallels between ideal free distribution of animal species and the distribution of units of culture?

Dan said...

Out of curiosity what 5 cultures are the ones you're talking about?

Wenyun said...

Those high cultural diversity places used to be relativily high wealth areas. What cause the wealth pattern shift?

Deepta said...

The Moore article mentions the higher language density along coastlines; is this primarily due to new groups arriving there? Is there any research looking at the languages to see if there is a possibility some are actually dialects of languages from other areas?

Justin Smith said...

Is cultural diversity a "good thing"?

Biogeographers see a strong negative trend between latitude and cultural diversity that they haven't fully explained. Economists see a strong positive trend between latitude and per-capita gross domestic product (wealth) that they haven't fully explained. A negative correlation between cultural diversity and wealth might be inferred, and may be tested for, though I'm not aware of any such studies.

Harvard political scientist Robert Putnam has recently documented (Scandinavian Political Studies, June 2007, 30:2, p. 137-174) that increased ethnic diversity leads to decreased levels of trust, altruism, reciprocity, cooperation, and community involvement. Cultural diversity goes hand in hand with ethnic diversity, so Putnam gives us a possible micro-level explanation for this macro-level pattern.

In other words, does cultural homogeneity in a society facilitate productivity and the generation of wealth? Does cultural diversity mean a society is more likely to be stuck in poverty?

Senorita Myra said...

Is it possible a study could be done on how recent trade laws between countries affects language and/or culture groups? I may be reading into it too much, but I get a sense from Collard and Foley that human distribution of food, between groups, has become a recent phenomenon not to be underestimated. Just recently I read an article on how fast languages were "dying" across the globe....What is the rate at which languages and/or cultures die historically? Has anyone looked at that? Can that be measured?

Michael said...

During empirical conquest, are languages and cultures assimilated at a predictable rate? or are there lurking variable (relative proximity on a language phylogeny, for example) that may account for different observable rates?

Locations of visitors to this page