Talk:Afro–Latin Americans/Archive 4

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Population figure, Brazil

Latest comment: 10 years ago46 comments4 people in discussion

I've replaced it with the Brazilian sources, the Brazilian census as informed by the census carried out by the government: ~6.84% black or 14,500,000 + 43.80% if including (multiracial) pardo or 83,540,920 (http://www.ibge.gov.br/english/estatistica/populacao/censo2010/caracteristicas_da_populacao/tabelas_pdf/tab3.pdf)

"Black Brazilians" are for the most part 'multiracial' with a greater African contribution than average. "Pardos" are more balanced mixed, and in some parts of Brazil more significantly Native American in ancestry. "Whites" usually are mixed too, but with a greater degree of european ancestry.

Genetic studies describe Brazil even better since they show the actual ancestry of the Brazilian population, hence I also posted this link which shows the degree of ancestry in the whole Brazilian population after averaging 'black', 'pardo' and 'white' ancestries according to their proportions.: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0017063

International media report often label pardo as 'black'. Though they have large African ancestry, their non African ancestry often tends to be even larger, so reports like Brazil has the second largest African population after Nigeria or 'blacks' are majority in Brazil may lead to the wrong conclusion that African ancestry is the most important in Brazil, when it is not, even though it is large and important there is even larger European ancestry as well as Native American input.Grenzer22 (talk) 15:41, 16 January 2014 (UTC)

Please read the preceding discussion, the figure for Brazil was discussed in detail there.

To recap, this article is about African ancestry, not self-identification as "black" (or "brown"), and this is something the Brazilian census doesn't measure directly. The colour-based self-identification is a good yardstick, but as the study you cited shows, some people who choose "brown" on the census have no African ancestry so we can't use the "brown" figure as an accurate representation of African ancestry. The genetic study itself doesn't give us a useful figure for the pop box either, because it only gives us a percentage of African ancestry within the population, not a percentage of individuals with African ancestry (when it says "x% African" this means the average person has x% African ancestry, not that x% of the population has African ancestry). We found a wide range of sources saying that approximately half the Brazilian population has African ancestry including statements by politicians and UN representatives. We've used this estimate to come up with the final 100M estimate (~half of ~200M).

I have reverted your change as it goes against the latest consensus here on the talk page, and also because it presents the data in a completely different format to the rest of the pop box. I'm happy to hear your arguments and see any other sources you can find, but please leave the table as it is until you get agreement here to change it.

Tobus (talk) 03:31, 17 January 2014 (UTC)

The figure I posted comes from the Brazilian census. ~44% of Brazilians identified themselves as 'pardo' not as 'black'. They are african descendants, and they are mentioned in the figure I gave, from the Brazilian government, by they are not africans, they are actually on average overwhelmingly non african in ancestry. Just read the section on the pardo. Genetic studies have proven it time and again.Grenzer22 (talk) 15:39, 17 January 2014 (UTC)

Just to add the figure I gave comes from the answers Brazilians themselves have given to the census. That should come first and foremost. I added a genetic link, since it shows the actual ancestry of Brazilians.

International media report often label pardo as 'black'. Though they have large African ancestry, their non African ancestry often tends to be even larger, so reports like Brazil has the second largest African population after Nigeria or 'blacks' are majority in Brazil may lead to the wrong conclusion that African ancestry is the most important in Brazil, when it is not, even though it is large and important there is even larger European ancestry as well as Native American input.Grenzer22 (talk) 15:43, 17 January 2014 (UTC)

Genetic studies have shown the Brazilian population as a whole to have European, African and Native Americans components. The European ancestry is greater among "white" and "pardo" Brazilians, the African ancestry is greater among "black" Brazilians, and Native American input is present throughout though to a lower degree generally.

An autosomal study from 2013, with nearly 1300 samples from all of the Brazilian regions, found a pred. degree of European ancestry combined with African and Native American contributions, in varying degrees. 'Following an increasing North to South gradient, European ancestry was the most prevalent in all urban populations (with values up to 74%). The populations in the North consisted of a significant proportion of Native American ancestry that was about two times higher than the African contribution. Conversely, in the Northeast, Center-West and Southeast, African ancestry was the second most prevalent. At an intrapopulation level, all urban populations were highly admixed, and most of the variation in ancestry proportions was observed between individuals within each population rather than among population'.^[1]

Region	European	African	Native American
North Region	51%	17%	32%
Northeast Region	56%	28%	16%
Central-West Region	58%	26%	16%
Southeast Region	61%	27%	12%
South Region	74%	15%	11%

A recent autosomal DNA study (2011), with nearly 1000 samples from all over the country ("whites", "pardos" and "blacks"), according to their respective proportions, found out a major European contribution, followed by a high African contribution and an important Native American component. http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0017063 "In all regions studied, the European ancestry was predominant, with proportions ranging from 60.6% in the Northeast to 77.7% in the South". http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0017063 The 2011 autosomal study samples came from blood donors (the lowest classes constitute the great majority of blood donors in Brazil. http://www.amigodoador.com.br/estatisticas.html Profile of the Brazilian blood donor=, and also public health institutions personnel and health students. The study showed that Brazilians from different regions are more homogenous than previously thought by some based on the census alone. "Brazilian homogeneity is, therefore, a lot greater between Brazilian regions than within Brazilian regions".^[2]

Region^[3]	European	African	Native American
Northern Brazil	68,80%	10,50%	18,50%
Northeast of Brazil	60,10%	29,30%	8,90%
Southeast Brazil	74,20%	17,30%	7,30%
Southern Brazil	79,50%	10,30%	9,40%

According to an autosomal DNA study from 2010, the composition of Brazil would be about 77,1% european, 14,3% ssa african and 8,5% native american. It showed also that physical indicators such as skin colour, colour of the eyes and colour of the hair did not correlate perfectly well with the genetic ancestry of each person http://www1.folha.uol.com.br/folha/ciencia/ult306u633465 http://www.ncbi.nlm.nih.gov/pubmed/19639555

It is important to note that "the samples came from free of charge paternity test takers, thus as the researchers made it explicit: "the paternity tests were free of charge, the population samples involved people of variable socioeconomic strata, although likely to be leaning slightly towards the ‘‘pardo’’ group".http://onlinelibrary.wiley.com

Region^[4]	European	African	Native American
North Region	71,10%	18,20%	10,70%
Northeast Region	77,40%	13,60%	8,90%
Central-West Region	65,90%	18,70%	11,80%
Southeast Region	79,90%	14,10%	6,10%
South Region	87,70%	7,70%	5,20%

An autosomal DNA study from 2009 found a similar profile. http://www.ncbi.nlm.nih.gov/pubmed/20129458

Region^[5]	European	African	Native American
North Region	60,6%	21,3%	18,1%
Northeast Region	66,7%	23,3%	10,0%
Central-West Region	66,3%	21,7%	12,0%
Southeast Region	60.7%	32.0%	7.3%
South Region	81,5%	9,3%	9,2%

According to another autosomal DNA study from 2008, by the University of Brasília (UnB), European ancestry dominates in the whole of Brazil (in all regions), accounting for 65,90% of heritage of the population, followed by the African contribution (24,80%) and the Native American (9,3%).http://bdtd.bce.unb.br/tedesimplificado/tde_busca/arquivo.php?codArquivo=3873 Grenzer22 (talk) 15:50, 17 January 2014 (UTC)

Just to add the source I gave is the one from the census undertaken by the Brazilian government, the IBGE:

[quote]The Brazilian Institute of Geography and Statistics or IBGE (Portuguese: Instituto Brasileiro de Geografia e Estatística), is the agency responsible for statistical, geographic, cartographic, geodetic and environmental information in Brazil. The IBGE performs a national census every ten years, and the questionnaires account for information such as age, household income, literacy, education, occupation and hygiene levels.[quote]

This is the link I gave: http://www.ibge.gov.br/english/estatistica/populacao/censo2010/caracteristicas_da_populacao/tabelas_pdf/tab3.pdf

The other link is that of thorought autosomal genetic study with "white", [[[pardo]] and "black" Brazilians from all over the country, according to their respective proportions. http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0017063 Grenzer22 (talk) 17:23, 17 January 2014 (UTC)

Just to sum it up. The figure I gave should be added as:

It is a reliable source from the Brazilian official census itself, based on self identification of Brazilians themselves. Pardo Brazilians have been included if including any degree of African ancestry.

To add in half as African Brazilians would imply that half Brazilians would be just like Africans or hint at half of Brazilians being pred. African in ancestry, which goes against all genetic studies covering the whole of Brazilian population being undertaken so far.

The figure I gave is about the same as that former one provided, but the depiction is more accurate, since it makes the distinction between pardo and black Brazilians, but at the same time it is hinted that both groups could be lumped together if any degree of significant ancestry is taken into account, regardless of the predominant genetic ancestry.Grenzer22 (talk) 15:58, 17 January 2014 (UTC)

Firstly, as per WP:BRD, when one of your edits is contested please leave the page as it was before your edit until the discussion reaches a conclusion.

The problem here seems to be one of definition. You've provided the Brazilian census as a source, which says how many people consider themselves "preto" or "pardo" ("black" or "brown"). You've also provided genetic studies which say how much of the average Brazilian DNA is African. The pop box however is showing neither of these, what it's showing is "Estimated Population with African Ancestry".... if the box were called "Population of Black or Brown People" then we'd use the census, if it were "Average Proportion of African Ancestry" then we'd use the genetic studies. Since it's neither of those, we use estimations of the number of individuals with African ancestry instead.

To sum it up from my perspective:

The Brazilian census says how many people consider themselves "black" or "brown", it does not say how many people have African ancestry.
Having African ancestry does not mean being "just like Africans", or being "predominately" of African ancestry... it simply means what it says, having some degree of African ancestry.
The distinction between "black" and "brown" ("pardo") is not relevant to the pop box, we're after the number of people with African ancestry, not the number of people who consider themselves a particular colour.

This same criteria has been applied to all the other population estimates in the box, so it makes no sense to treat Brazil differently. If you think the box should be displaying something different to what it does, then that's a different argument altogether as this would apply to all the other countries listed, not just Brazil.

Tobus (talk) 22:23, 17 January 2014 (UTC)

I have used the Brazilian census, as I said, just look it up. The source is the Brazilian government itself, based on what Brazilians themselves reported to the government. Obviously, this should be the primary information.

The edit was made on the information which was given before, and this is the version I am trying to keep. Just look at the history of the topic.

Since I have provided the link with the official information provided by the Brazilian government, I think it is fair to say it should be kept.Grenzer22 (talk) 00:22, 18 January 2014 (UTC)

I know your source is the Brazilian census, and you have accurately reported what it says. The problem is that it just does not say how many people there are in Brazil with African ancestry. The "black" figure is way too low - most people with African ancestry don't consider themselves "black". The "brown" figure is too high - some people who call themselves "brown" are mixed Amerindian/European and have no African ancestry. You also don't include those people with African ancestry who consider themselves "white". Your source is reliable and your text represents it accurately, but you've answered the wrong question.

The source I used in the article is from Mireille Fanon-Mendes-France, a member of the UN panel on Afro-descendants reporting on a 10-day mission in Brazil("Afro-Brazilians represent more than half of the Brazilian population"[1]). A similar figure has also been stated publicly by Brazilian state prosecutor Christiano Jorge Santos ("nearly half of the Brazilian population identify with being Afro-Brazilian"[2]) as well as Brazilian economist and researcher Marcelo Neri ("there are 97 million African descendants in Brazil, comprising 50.6 percent of the Brazilian population"[3]). There are also lots of similar estimations from world media ([4][5][6][7] etc.) These are all in line with the estimation from the Minority Rights Group who regularly report on minority groups to both governments and the UN ("It is estimated that between 65 million and 120 million Brazilians are of African ancestry"[8]). These are all reliable sources which directly state an estimation of African ancestry, and we should use them instead of WP:SYNTHing figures from the colour-based groupings of the census.

Tobus (talk) 02:41, 18 January 2014 (UTC)

After having read all this, i have to say that the numbers given by Brazil's government are the ones that must be used on this article, the problem in this is that one editor wants to include "pardo" altogether on the afro population, but that is a classification apart created by Brazil's government, and it might include people that don't has a majoritarily african heritage, and can't be afro. It even has it's own separated article Pardo, and i don't see that article including afro people on the official number. To include that classification on an article for afro people is wrong. The best is to only keep afro here and maybe an aclaration pointing that pardos might have african heritage aswell. Prestonishere (talk) 06:33, 18 January 2014 (UTC)

Thanks for your input Preston. You seem to be saying that "Afro-" is synonym for "black" on the census - meaning full or majority African ancestry. This seems at odds with the page as it stands now, which includes people with non-predominant African ancestry (such as Nicole Richie) as examples, and also with the way the term is used in various sources and media. I think this disagreement about what to use for the Brazilian population stems from a general lack of agreement as to what this page is about... is this page about "Black" Latin-Americans (as defined by their own self-identity and culture), or is it about Latin Americans who have African ancestors (as defined by their genetics/family tree)? Given the current content of the page and the way the term "Afro-" is used in the sources and media, I have assumed the latter but it seems you have a different interpretation. Tobus (talk) 08:49, 18 January 2014 (UTC)

The 'if' when referring to pardo Brazilians already makes clear that many of them could fall in the group in question. The reader would then proceed to the pardo topic and learn more about this segment of the Brazilian population.

I believe the information as it is should be kept since it is the official Brazilian government information, based on what Brazilians themselves declared to the census.Grenzer22 (talk) 15:21, 18 January 2014 (UTC)

Just to add that no other institution has carried a census of the entire population, and based on what Brazilians themselves declared or identified as, but the Brazilian government itself. All other figures are mere estimates.Grenzer22 (talk) 15:28, 18 January 2014 (UTC)

Tobus, as for your links, as I said, international media (and some Brazilian sources even) reports often label the pardo as 'black' (that's how they get the 50% figure, adding the ~7% self identified 'blacks' to the 43% self identified 'pardos'). Reports like Brazil has the second largest African population after Nigeria or 'blacks' (like one can find in some of the links you quoted) are majority in Brazil can be highly misleading in the depiction of the country, since what 43% of Brazilians have identifies themselves was as pardo not as blacks.

As a matter of fact, even though African ancestry is large and important there is even larger European ancestry as well as Native American input.

Anyway, if it is about having any degree of African ancestry, then the number of Americans certainly would be much higher, since according to various sources quite a good number of white Americans have african ancestry themselves, not to mention 'latinos' who, either coming from Mexico, the Caribbean or South America, tend, in their majority, to have at least some degree of ancestry. The average Latin American, be him the average Mexican, the average hispanic Caribbean, the average Colombian, or the average Venezuelan, f.e, have at least some degree of African ancestry, according to genetic studies.

It wouldn't make sense either to have a separate 'coloured' vs 'black' South African category, since both groups also have African ancestry. But that's not the case here nor anywhere else in the media when it comes to South Africa.

Basically only Brazil gets this different treatment, and exactly because of that the way in which Brazilians themselves have identified to the census should be respected (as reported by the Brazilian government itself, the only institution to carry out this survey encompassing all Brazilians). Even more, the overwhelming majority of genetic studies show clearly the majority of Brazilians have a pred. degree of non african ancestry (even if they also tend to have at least some degree of ancestry, like the majority of Latin Americans, according to the genetic studies).

Therefore, the detailed answer x% of black and y% of pardo provides a more accurate representation of Brazil. One can go to the pardo topic on wiki, read about them there, and check their ancestral profile, their distribution in the country, etc, and make their own conclusions about it. At the same time, one will not get a narrow view on what would be the African impact in Brazil, since pardo via the 'if' could also be included; and in the section below on Brazil, on the same article, a general view of the Brazilian population, when it comes to actual ancestry, is reported.Grenzer22 (talk) 16:41, 18 January 2014 (UTC)

Sorry, Grenzer but you've completely missed my main point - the Brazilian census simply does not say how many Afro-Brazilians there are - all it says is "black", "white" and "pardo" which is not the same thing. All 3 categories include people with African ancestry, and all 3 of them include people who don't. It's impossible to get an accurate estimate of African ancestry from those census categories. On the other hand we have a wide range of sources from all over the world that give a direct estimate of African ancestry, in some cases saying "Afro-Brazilians" directly. Yes, some of these use the US-interpretation of "black", but that's because they're aimed at a US audience where "black" and "Afro-" are interchangeable and mean any degree of African ancestry, unlike in Latin America where "black" means full or predominant African ancestry and "Afro-" isn't used much at all.

While I agree to some degree with much of what you have said, there are a number of misunderstandings:

Not one single Brazilian said they were "Afro-" on the census, it wasn't available a category. To come up with our own estimate of "Afro-" individuals is not allowed as per WP:SYNTH.
American "whites" have very little African ancestry (<2%) whereas Brazilian "whites" have ~30% African ancestry on average.[9]
Most Latin American countries get the same treatment as Brazil and most have distinct "Black" and "Brown" categories on the census. We have treated them all countries the same here and used independent estimations of ancestry instead of trying to calculate it based on the census colour categories. Have a look at the page and notice how your edit sticks out as completely different to the rest... it's you who is treating Brazil differently.
"x% black and y% pardo" may be a better representation of racial identity in Brazil, but this page isn't about racial identity, it's about African ancestry. Since neither "black" nor "pardo" directly correlates with African ancestry, using your "x + y" is a less accurate representation of what this page is about.

I sympathise with where you're coming from, and strongly believe that self-identified "race" is more important than a purely genetic definition. But in this case "Afro-xxx" as used in this page is not a racial term, but a genetic/ancestry-based one. This term is not widely used in Latin-America and the concept itself is only recent (as opposed to say "mestizaje" or "blanqueamiento") - which is why there is no option to self-identify with "Afro-" in the Brazilian census. We can try to fit the "black", "pardo" and "white" categories into the essentially foreign "Afro-" construct but it's not going to give us an accurate answer - they are defining different things.

Tobus (talk) 04:21, 19 January 2014 (UTC)

As I said, self identification is key here, as reported to the government. The category black has historically being associated with the Brazilian population of greater African ancestry (from colonial times to now). The Brazilian population, even those who identified as black to the census, tends to have european, african and native american ancestries. Usually, non african ancestry is predominant in the majority of Brazilians, particularly with the white Brazilian and pardo Brazilian segments, according to the vast majority of genetic studies. If it is going by a single ancestral component to define it, then one could also say that European Brazilians would account for say 90% of the population, since most Brazilians also have European ancestry. Not to mention Native American ancestry.

Anyway, if it is about having any degree of African ancestry, then the number of Americans certainly would higher, since according to various sources quite a good number of white Americans have african ancestry themselves, about 1/3 of them, according to a genetic study (going as high as 20% in some individuals) even if the average is low: http://www.upi.com/Top_News/2002/05/08/Analysis-White-prof-finds-hes-not-2/UPI-53561020909970/; http://essays.backintyme.biz/item/5. Other studies have reached similar conclusions: http://www.disabled-world.com/health/cancer/racial-disparities.php.

Not to mention latino Americans who, either coming from Mexico, the Caribbean or South America, in their majority, do tend to have at least some degree of ancestry. The average Latin American, be him the average Mexican, the average hispanic Caribbean, the average Colombian, or the average Venezuelan, f.e, have at least some degree of African ancestry, according to genetic studies.

It wouldn't make sense either to have a separate coloured South African vs black South African categories, since both groups also have African ancestry (incidentally White South Africans of Dutch ancestry - Afrikaner South Africans - also tend to have African ancestry on average - http://repository.up.ac.za/bitstream/handle/2263/5168/Greeff_Deconstructing%282007%29.pdf?sequence=1 - even though historically they have never been grouped with neither black South Africans nor with the coloured South African category).

Basically only Brazil gets this different treatment, and exactly because of that the way in which Brazilians themselves have identified to the census should be respected (as reported by the Brazilian government itself, the only institution to carry out this survey encompassing all Brazilians).

Even more, as I said, the overwhelming majority of genetic studies show clearly the majority of Brazilians have a pred. degree of non african ancestry (even if they also tend to have at least some degree of ancestry, like the majority of Latin Americans, according to the genetic studies).

Therefore, the detailed answer x% of black and y% of pardo provides a more accurate representation of Brazil. One can go to the pardo topic on wiki, read about them there, and check their ancestral profile, their distribution in the country, etc, and make their own conclusions about it. At the same time, one will not get a narrow view on what would be the African impact in Brazil, since pardo via the 'if' could also be included; and in the section below on this article in discussion, a general view of the Brazilian population, when it comes to actual ancestry, is reported (including white Brazilians).Grenzer22 (talk) 15:08, 19 January 2014 (UTC)

You've ignored my points and have pretty much just cut and pasted your previous comments. In the interests of resolving this, can you please take the time to read and respond to the following points:

You state that "self identification is key here, as reported to the government" but the Brazilian census doesn't give people the option to self-identify as "Afro-Brazilian". What is the reasoning behind such a statement?
You state that "if it is about having any degree of African ancestry, then the number of Americans certainly would higher", but the US census has separate categories for "white" and "black" Latino/Hispanic and also allows for the selection of multiple categories - so the figure already includes the groups you mention. If you have a reliable WP:RS that gives an alternative figure then feel free to update it. (By my reading of the 2010 census[10] the figure here is massively overstated - it should be 1.2M/0.4% from the "HISPANIC OR LATINO AND RACE: Black or African American alone" figure... this table[11] gives 1.4M including Black+other mixed-race Latino/Hispanic.)
You state "It wouldn't make sense either to have a separate coloured South African vs black South African categories" but this is exactly what you are doing - separating Afro-Brazilians into "pardo" and "black" categories... I don't understand, what is the point you are trying to make here?
You state "Basically only Brazil gets this different treatment" but the edit you made is the one that is treating Brazil differently... what is your reason for presenting the "Afro-" population in Brazil differently to the "Afro-" population in other Latin American countries?
You say "the overwhelming majority of genetic studies show clearly the majority of Brazilians have a pred. degree of non african ancestry" which is not under dispute, but I fail to see how it is relevant - are you saying that "Afro-" means having a "pedominant degree of African ancestry"?
You insist that "x% of black and y% of pardo provides a more accurate representation of Brazil" but there is no direct correlation between these groups and African ancestry. Why do you think it is important that we split Afro-Brazilians into different categories based on skin colour?

Tobus (talk) 09:00, 20 January 2014 (UTC)

Again, you are ignoring my points.

The category black has historically being associated with the Brazilian population of greater African ancestry (from colonial times to now). Historically it was applied to no one else but the African population in Brazil. The pardo are more multiracial in outlook, and a brief reading on the wiki topic on them will show it to you. In spite of that, an option for an inclusion of pardo Brazilians as Afro Latin Americans is given on the edit I'm trying to save. The Brazilian population, however, and as a matter of fact, even those who identified as black to the census, tends to have European, African and Native american ancestries.

The reason for keeping the census information is that that's how Brazilian themselves, in their entirety, reported themselves. No one else did it, and it is official information. One can read about these categories and check their ancestral profiles, something that even on this article has been done, and make their conclusions. No other census has been carried out, and most of the information elsewhere comes from estimates, some quite misleading, as in lumping black Brazilian and pardo Brazilian as one category and implying that half of Brazilians would be like Nigerians or African Americans, which is quite a misrepresentation of Brazil.

The census does have both black and pardo options, and these two groups differ in their ancestral profile remarkably. Black Brazilians are the ones who usually tend to be pred. african in ancestry, not pardo Brazilians. By lumping them like if they were just the same is misrepresenting Brazil. On other hand showing the census distinction underlines the details of the composition of the Brazilian population, not leting it to be misrepresented.

Non african ancestry is predominant in the majority of Brazilians, particularly with the white Brazilian and pardo Brazilian segments, according to the vast majority of genetic studies. If it is going by a single ancestral component to define it, then one could also say that European Brazilians would account for say 90% of the population, since most Brazilians also have European ancestry. Not to mention Native American ancestry.

If it is about any degree of African ancestry, then we'll have to change the numbers for all Latin American countries, and the figure will rise up to 500 million. By neglecting the average degree of some African ancestry in the average Latin American, and not doing it with Brazil, one will be not coherent.

In international media the categories of South Africa are always respected. They don't lump coloured South African and black South African categories as the same, even if both groups do show African ancestry (incidentally White South Africans of Dutch ancestry - Afrikaner South Africans - also tend to have African ancestry on average - http://repository.up.ac.za/bitstream/handle/2263/5168/Greeff_Deconstructing%282007%29.pdf?sequence=1 - even though historically they have never been grouped with neither black South Africans nor with the coloured South African category).

It is not done with Latin Americans, who on average do exhibit some degree of African ancestry, it is not done with coloured and Afrikaner South Africans, why should it be with Brazilians?

As I said, basically only Brazil gets this different treatment, and exactly because of that the way in which Brazilians themselves have identified to the census should be respected (as reported by the Brazilian government itself, the only institution to carry out this survey encompassing all Brazilians).Grenzer22 (talk) 11:59, 20 January 2014 (UTC)

Thanks for replying, I can understand where you're coming from a bit better now.

While I commend you for wanting to explain the racial identity of Brazil in detail, I have to ask you if this pop box if the appropriate place to do it. It's explained in much greater detail in various places, such as the Brazil section of this page, the Black People page, Demographics of Brazil etc. and of course in the Brazil page itself which is linked to from the pop box. This page is about a different racial grouping than is commonly used in Latin America, I have no problem at all with country-specific breakdowns being explained in detail in the page body, but I don't think it's necessary, appropriate or consistent to do it for one country in a pop box for "African Ancestry".

Some of the reasons I don't think it's a good idea to separate the Afro-Brazilian population into colour-based racial groups in the pop box are:

"Afro-Brazilian" is not part of the Brazilian self-identity, it's an introduced term that uses a different system of racial grouping not found in the official census.
Brazilian government employees, the UN, Latin American documentary makers and the worldwide media all use "Afro-Brazilian" as an analog to "African-American" - any amount of African ancestry with no separation by physical appearance or degree of ancestry. I have yet to see any source talk about "black Afro-Brazilians" and "pardo Afro-Brazilians". The two classification systems are independent and can't be used together.
Using a color-based division artificially creates two separate classes of "Afro-"... this page is specifically about a concept that spans the color groupings on the census, so quoting two separate figures based on colour is counter-intuitive.
It's extremely clumsy looking and, for the topic of the pop box ("Estimated Population with African Ancestry"), unnecessary

I now understand and accept your point about South African being treated differently. I expect it's because "black" South Africans are the vast majority (80% "black", 10% "coloured" and 10% "white") so the difference between "Afro-" and "black" is very small, whereas "black" Latin Americans are a minority in most countries and the "black" and "Afro-" figures are very different. I will have to point out though that you are wrong about Brazil being singled out in this way, virtually all Latin American countries are in the same situation - for example the Venezuelan census reports 3% "black" and 50% "moreno" and we are using a single "Afro-" estimate for the pop box. The same goes for nearly all the other countries - they have a "black"/"mixed" distinction but we use a single ancestry-based estimate for "Afro-" in the popbox (usually not from the census) and explain each country's particular racial breakdown where appropriate in the page body.

Considering this, are you willing to accept using a single estimate for Brazilians with African ancestry in the pop box and then expanding on this if needed in the Brazil section of the page? This gives a clean simple pop box with a figure that directly relates to the page topic, but still allows for a clear explanation of the unique demography and racial groupings of each country.

Tobus (talk) 00:30, 21 January 2014 (UTC)

Again, as I said, if you look at our history, you'll see the africans brought here and those born in Brazil were called black, a category which is in the census. If it is about any degree of african ancestry, we'll have to change the figures of other Latin American countries, since genetic studies show the average Latin American to have some degree of african ancestry. As I said, only the IBGE (government agency) carries out the census, and the answers to the census should be respected. One can go to the sections and learn more about them, their ancestral profile, where it is shown that african ancestry is widespread in Brazil, being something rather difficult to establish clearly what is and what is not african Brazilian based on ancestry. I think the explanations on the Brazilian section of the Afro Latin American topic, on the wikipedia article on pardo, black and white Brazilians have information enough concerning the ancestry of Brazilians and the african ancestry in Brazilians.

The way it is, with the census categories, it is better IMO, since that's how Brazilians reported themselves. It won't alter the estimation, since the figure according to the census would be ~97 million, which is the same as the previous ~ 100 million. It is just that usually, as I said, Brazil is misrepresented. Look at this the Guardian article for example. "Preliminary results show 50.7% of Brazilians now define themselves as black or mixed race compared with 47.7% whites [...] Preliminary results from the 2010 census, released on Wednesday, show that 97 million Brazilians, or 50.7% of the population, now define themselves as black or mixed race, compared with 91 million or 47.7% who label themselves white. [...] But the proportion of people declaring themselves black or mixed race has risen from 44.7% to 50.7%, making African-Brazilians the official majority for the first time". (http://www.theguardian.com/world/2011/nov/17/brazil-census-african-brazilians-majority) They don't explain that in fact white, pardo and black Brazilians, ancestrally, are actually mixed. They label pardo and black Brazilians as the same, they ignore the (overwhelming) non african ancestry of pardo ('mixed race' Brazilians), and they fail to mention that pardo Brazilians are 43% of the population and black Brazilians are 7%. Actually, they place black ahead of pardo Brazilians in the sentences, as if they comprised a higher % of the population, but in fact they comprise a mucher lower % of the population.

As I said, the reason for keeping the census information is that that's how Brazilian themselves, in their entirety, reported themselves. No one else did it, and it is official information. One can read about these categories and check their ancestral profiles, something that even on this article has been done, and make their conclusions. No other census has been carried out, and most of the information elsewhere comes from estimates, some quite misleading, as in lumping black Brazilian and pardo Brazilian as one category and implying that half of Brazilians would be like Nigerians or African Americans, which is quite a misrepresentation of Brazil.

So, keeping as it is now won't alter the numbers, but at the same time it will be more detailed and accurate, avoiding the misleading portrayals Brazil has been the target so often.Grenzer22 (talk) 11:24, 21 January 2014 (UTC)

Tobus, I have some sources that can be useful for you. This is a resource from IBGE, which is responsible for official Brazilian census. Page 41, 7.8% reported their race as "negro" (black) and 1.4% as "preto" (another word for "black", a total of 9.2%). However, on page 49, 11.8% of people reported that their family is of African origin. When multiple answers were allowed (page 52), 25.1% of the respondents claimed to be "Afro-descendentes" (of African descent) and 28.8% as having a "Negro" (Black) ancestry.

Notice (page 50) that only 39.5% of people who claimed to be of "negra" (Black) color said that they have an African ancestry. Remember that there is a lot of self-hating and low self-steem people in Brazil who deny their African ancestry so, of course, these figures may be far from reality, but this is what we have by now. The "whiten ideology" and self-hating can also be discussed in the article.

I'd like to add that the official Brazilian census does count "Preto" (Black) and "Pardo" (Brown) people together as "Black population".[12] If people like ir ot not, this is another discussion. Moreover, Brazilian census have some issues, since, in many cases, it is not the person who classify their own race, but a relative. I was alone at home when the census guy arrived and I was the one who racially classified my entire family. This can bring mistakes, since a person may consider her/himself Black, but a relative classifies them as Brown or even White. There are some studies that show that Brazilians tend to "whiten" people when they have to classify the race of others. This may have some influence in the census. Then, Brazilian census is not completly accurate, not only because many Brazilians deny their Black heritage, but also because, in many cases, a single person of the family classify the race of everybody else who live in the house. Xuxo (talk) 20:23, 21 January 2014 (UTC)

No, the IBGE does not count them as black at all. Otherwise, the two categories would not exist. And they do exist, pardo and black. One of your sources is a news source, and the other one is not lumping pardo Brazilians as black Brazilians at all. It is developing the subject, which is completely different. Anyway, in the option I'm trying to save the pardos could be included and are mentioned.

Genetically, pardo Brazilians from various places, according to a genetic study from 2011, are for the most part pred. non african in ancestry, definitely in most of Brazil (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3040205/):

Pardo Brazilians from Ilhéus: ~70% non african genetically

Pardo Brazilians from Belém: ~90% non african genetically

Pardo Brazilians from Fortaleza: ~85% non african genetically

Pardo Brazilians from Porto Alegre: ~65% non african genetically

Incidentally, the most pardo part of Brazil, Northern Brazil, is one of the least african genetically.

The same study mentioned (with nearly almost 1000 samples from all over the country, "whites", "pardos" and "blacks", according to their proportions) has also concluded that European ancestry is the predominant ancestry in Brazil, accounting for nearly 70% of the ancestry of the population. "In all regions studied, the European ancestry was predominant, with proportions ranging from 60.6% in the Northeast to 77.7% in the South".

The 2011 autosomal study samples came from blood donors (the lowest classes constitute the great majority of blood donors in Brazil (http://www.amigodoador.com.br/estatisticas.html), and also public health institutions personnel and health students.

Genomic ancestry of individuals in Porto Alegre Sérgio Pena et al. 2011 .^[6]
colour	amerindian	African	European
white	9.3%	5.3%	85.5%
pardo	11.4%	44.4%	44.2%
black	11%	45.9%	43.1%
total	9.6%	12.7%	77.7%

Genomic ancestry of individuals in Ilhéus Sérgio Pena et al. 2011 .^[6]
colour	Amerindian	African	European
white	8.8%	24.4%	66.8%
pardo	8.9%	30.8%	60.3%
black	10.1%	35.9%	53.9%
total	9.1%	30.3%	60.6%

Genomic ancestry of individuals in Belém Sérgio Pena et al. 2011 .^[6]
colour	Amerindian	African	European
white	14.1%	7.7%	78.2%
pardo	20.9%	10.6%	68.6%
black	20.1%	27.5%	52.4%
total	19.4%	10.9%	69.7%

Genomic ancestry of individuals in Fortaleza Sérgio Pena et al. 2011 .^[6]
colour	Amerindian	African	European
white	10.9%	13.3%	75.8%
pardo	12.8%	14.4%	72.8%
black	N.S.	N.S.	N.S

Grenzer22 (talk) 22:28, 21 January 2014 (UTC)

Sérgio Pena, the most renowned Brazilian geneticist: "the autosomal genetic analysis that we have performed in non-related individuals from Rio de Janeiro shows that it does not make any sense to put "pretos" and "pardos" in the same category".http://www.laboratoriogene.com.br/geneImprensa/2009/pensamento.pdf

IBGE specialist, José Luiz Petruccelli: "O especialista diz não ser correto, para efeito de pesquisas, reunir pardos e pretos em um só grupo, de negros". "It is not correct to lump, in the researches, pardo and black Brazilians as one group".http://noticias.terra.com.br/educacao/voce-sabia/qual-a-diferenca-entre-preto-pardo-e-negro,395c952757b7e310VgnVCM5000009ccceb0aRCRD.html

White skinned Brazilians have classified themselves as pardo in some places, as IBGE worker José Luiz Petrucelli has explained: "Ele ainda conta uma história curiosa sobre a situação no Distrito Federal. "A população local, por mais branca que seja a sua pele, se classifica como parda porque vê os brancos como os funcionários públicos que vieram de fora". "The local population in the Distrito Federal, no matter how white its skin, tends to classify itself as pardo since it sees whites as public servants who came from outside". http://noticias.terra.com.br/educacao/voce-sabia/qual-a-diferenca-entre-preto-pardo-e-negro,395c952757b7e310VgnVCM5000009ccceb0aRCRD.html

Actually, a genetic study addressing self reported and actual (real) ancestry verified that European ancestry, not African ancestry, it the most underestimated one, particularly among pardo Brazilians. http://www.meionews.com.br/index.php/noticias/21-estado-do-rio/4607-negros-e-pardos-do-rio-tem-mais-genes-europeus-do-que-imaginam-segundo-estudo.html

The Nação Mestiça movement (Pardo Brazilians movement) is against it, as they say it in their page:

"Pardo. Pardo é um termo referente a pessoas mestiças de cor entre branco e preto. É usado no Brasil, para classificação de cor/raça pelo Instituto Brasileiro de Geografia e Estatística (IBGE). No censo de 2.000, 38,5% dos brasileiros se auto-declararam como sendo pardos. A palavra pardo deriva do latim 'pardus', significando leopardo. Durante muitos anos, vem-se usando o termo pardo como grupo étnico do Brasil, mas o mesmo já foi substituído por mestiço no censo de 1890, retornando à expressão pardo no censo de 1940 e permanecendo até os dias atuais. Quem é pardo? O Brasil adota a auto-classificação para classificar a sua população em diferentes cores: branco, preto, pardo, amarelo e indígena. O termo pardo é mais antigo que o próprio Brasil: na carta de Pero Vaz de Caminha, durante a chegada dos portugueses ao Brasil, em 1500, ele relatou ao rei de Portugal que os indígenas brasileiros eram "pardos, um tanto avermelhados, de bons rostos e bons narizes, bem feitos".

De uma maneira sucinta, a maior parte dos brasileiros que se classificam como pardos usam o mesmo critério daqueles que se classificavam como mestiços nos censos antigos: são pessoas de ascendência mestiça, frutos de quinhentos anos de miscigenação entre índios, brancos e negros."

Pardo não é sinônimo de negro ou de afrodescendente. Declarar-se pardo não significa, assim, declarar-se afrodescendente, ou seja, descendente de africanos, nem negro." http://www.nacaomestica.org/pardo.htm

"Pardo and black or african descendant are not the same thing". Nação mestiça movement http://www.nacaomestica.org/pardo.htm

In spite of it, and considering that many pardo Brazilians may indeed have substantial african admixture, and given that some advocate the inclusion of them as black Brazilians, the option for the inclusion of pardo Brazilians in my edit has been saved.

One can read about these categories and check their ancestral profiles, something that even on this article has been done, and make their conclusions. No other census has been carried out, and most of the information elsewhere comes from estimates, some quite misleading, as in lumping black Brazilian and pardo Brazilian as one category and implying that half of Brazilians would be like Nigerians or African Americans, which is quite a misrepresentation of Brazil.

So, keeping as it is now won't alter the numbers, but at the same time it will be more detailed and accurate, avoiding the misleading portrayals Brazil has been the target so often.Grenzer22 (talk) 22:48, 21 January 2014 (UTC)

Firstly, am I correct in saying we both agree that there are ~100M Brazilians with African ancestry - you just want to report it based on skin colour and I don't?

Your argument for splitting is that it will be "more detailed and accurate", but you've provided a whole bunch of genetic studies that prove beyond any doubt that skin colour and ancestry aren't the same thing - some "white" Brazilians have African ancestry and some "pardo" and "black" Brazilians don't. Giving separate "black" and "pardo" figures would be very misleading and inaccurate because not all the African ancestry belongs to these two groups, and not everyone in these two groups has African ancestry. To be consistent with the genetic studies we'd have to give the "white" figure here as well as the "black" and "pardo" ones, something I think even you would find ridiculous.

I understand you have grown up with a very strict set of racial concepts and want that presented here, but remember this page is talking about a completely different concept, one that a colour-based interpretation doesn't fit neatly into. It's certainly within the scope of the article to discuss the colour-based concepts used in each country, but it's not appropriate or useful to use them in a population box called "Estimated Population with African Ancestry". What we want in that box is an estimate of how many people have African ancestry, and regardless of how Brazilians reported themselves in the census, the genetic studies all show that "black" and "pardo" are talking about something else entirely.

Tobus (talk) 11:56, 22 January 2014 (UTC)

The census figure should be kept (it is ~100 million anyway), but it is more detailed so that people can get more information by going to the other topics, and not be misled, as time and again, as I've showed, Brazil is quite incorrectly portrayed. Not to mention it is the official figure. I gave the link and one can check it there: http://www.ibge.gov.br/english/estatistica/populacao/censo2010/caracteristicas_da_populacao/tabelas_pdf/tab3.pdf.

As I said, if it is about any degree of African ancestry, we'll have change it entirely, to include the average Venezuelan, the average Colombian, in short, most Latin Americans, who according to genetic studies do have African ancestry. If the figure is replaced to include all of those who have any degree of ancestry, then we can change when it comes to Brazil. If not, it should as it is, with the official census information, and including pardo Brazilians, even if, as I have shown, their ancestry makes them, as a whole, quite different from Africans, and even if their inclusion as African Brazilians is highly controversial.

The way it is, it is better, since it highlights the contours of the Brazilian population. 100 million African Brazilians, without specifying that pardo Brazilians are being included is an incomplete and misleading information.Grenzer22 (talk) 13:08, 22 January 2014 (UTC)

You are entirely misrepresenting myself when you have said this: "I understand you have grown up with a very strict set of racial concepts and want that presented here".

To begin with, personally I don't believe in racial categories (these are just names made up during colonization, and if it were by myself they would have all been abolished long ago, without giving up protection of historically oppressed groups). But my personal opinion does not matter here, what I don't want is my country to be portrayed by other means than its population and the government present itself to be. Since usually the reasoning behind the ~100 million figure is to include black and pardo Brazilians as the same, as I've shown to you above with that the Guardian article above, I just made explicit by detailing the figure, and with the official information provided by the government.

I have not included any degree of African ancestry, because this has not been done with other Latin American countries, as I said. If it was, the vast majority of Colombians, Mexicans, Venezuelans, etc, should be included, because according to the vast majority of genetic studies all of them have at least some degree of genetic ancestry. And you can check it by yourself looking at individual genetic results on 23andme (wwww.23andme.com).

In short, I'm just detailing the figure so as not to mislead people, as often it is done.Grenzer22 (talk) 13:15, 22 January 2014 (UTC)

I'll repeat it again, so that perhaps you'll get it now. Any degree of African ancestry is not being used here, but a variety of sources, to each country. If it was any degree of ancestry, we would have to raise the figure to include, as I said, not only the majority of Brazilians, but also the majority of Colombians, Venezuelans, Mexicans, etc, who all tend to have at least some degree of ancestry, according to genetic studies on the subject (http://bdtd.bce.unb.br/tedesimplificado/tde_busca/arquivo.php?codArquivo=3873, page 33, as an example of a genetic study, one could quote many more, the majority show Latin Americans to have some degree of african ancestry, so the figure, if it was about any degree of African ancestry, should in fact be raised to around 500 million at least).

The way it is, with the census categories, it is better IMO, since that's how Brazilians reported themselves. It won't alter the estimation, since the figure according to the census would be ~97 million, which is the same as the previous ~ 100 million. It is just that usually, as I said, Brazil is misrepresented, they just place the figure adding black and pardo Brazilians, without explaining it. Look at this the Guardian article for example. "Preliminary results show 50.7% of Brazilians now define themselves as black or mixed race compared with 47.7% whites [...] Preliminary results from the 2010 census, released on Wednesday, show that 97 million Brazilians, or 50.7% of the population, now define themselves as black or mixed race, compared with 91 million or 47.7% who label themselves white. [...] But the proportion of people declaring themselves black or mixed race has risen from 44.7% to 50.7%, making African-Brazilians the official majority for the first time". (http://www.theguardian.com/world/2011/nov/17/brazil-census-african-brazilians-majority) They don't explain that in fact white, pardo and black Brazilians, ancestrally, are actually mixed. They label pardo and black Brazilians as the same, they ignore the (overwhelming) non african ancestry of pardo ('mixed race' Brazilians), and they fail to mention that pardo Brazilians are 43% of the population and black Brazilians are 7%. Actually, they place black ahead of pardo Brazilians in the sentences, as if they comprised a higher % of the population, but in fact they comprise a mucher lower % of the population. As I said, the reason for keeping the census information is that that's how Brazilian themselves, in their entirety, reported themselves. No one else did it, and it is official information. One can read about these categories and check their ancestral profiles, something that even on this article has been done, and make their conclusions. No other census has been carried out, and most of the information elsewhere comes from estimates, some quite misleading, as in lumping black Brazilian and pardo Brazilian as one category and implying that half of Brazilians would be like Nigerians or African Americans, which is quite a misrepresentation of Brazil.

So, keeping as it is now won't alter the numbers, but at the same time it will be more detailed and accurate, avoiding the misleading portrayals Brazil has been the target so often.Grenzer22 (talk) 13:35, 22 January 2014 (UTC)

I understand where you are coming from in not wanting the page to misrepresent things, but in this case I think you are overcompensating and have nothing to worry about. Just above the popbox, the first thing you see when you come to the article, is an image array of "Afro-Latin Americans", and quite clearly most of them are in the "pardo" category. When people see the "100M" Brazilians, they are going to think it's 100M people like in the images directly above... people like Nicole Richie, Adriana Lima, Rita Morena etc. There's no need to specify "pardo" in the popbox as it's already indicated by the images that "Afro-" doesn't mean "black". All splitting the figure does it does is make the popbox look clumsy and confuses the reader - instead of a simple (and as you agreed, accurate) "100M", they now have to add two figures together and read a bunch of text.... something they don't need to do for any of the other countries listed, and something they can do in lots more detail in lots of other places (including further down the page) if they wanted to.

I also note that your recent edit is doing a similar thing, over-explaining something that doesn't need explaining. The text says "Countries with significant black, mulatto, or zambo populations today include... ", there's no need for you to add "if including the pardo Brazilian population" because it's already been stated that it's including more than just the "black" population... plus the 100M figure doesn't come from including the "pardo" population, it comes from estimating African ancestry independently of any colour-based groupings.

The other main issue I have is that we cannot use the Brazilian census as a source because it doesn't give a figure for how many Brazilians have African ancestry. Adding "black" and "pardo" together is WP:SYNTH and is not allowed.

Tobus (talk) 10:33, 23 January 2014 (UTC)

I think for the sake of accuracy one should point out the "reasoning" behind the figure. And the "reasoning" as I just showed to you on the the Guardian article above is lumping black and pardo Brazilians (~43% + ~7% = ~50%). Often, as I showed, accompanied by quite a few misleading conclusions (such as that half of Brazil would be like Nigeria, etc). I haven't changed the figure. I pointed out how it came about (it does not come out of measuring African ancestry in Brazilians, or out of a census conducted on their own, but simply out of lumping black and pardo figures from the Brazilians census). As I said, if the article were really about any degree of African ancestry, we would have to raise the figure for all other Latin American countries. The average Colombian or Venezuelan, f.e, I repeat, has African ancestry. It is not like only 5 million out of 40 million Colombians have African ancestry and the rest does not. The same goes to the rest of Latin America.

I think the census link should be kept, since it is the official government information. I have checked the figure for Venezuela, and it also comes from the Venezuelan government. Why should it not be with Brazil? At least the way newspapers arrive at ~100 million figure is made explicit. And it should, if only for the sake of clarification.

I'm not trying to change the article, but to keep the description of Brazil as accurate as it can be. If I were in charge I would post the genetic studies saying that the vast majority of Latin Americans have African ancestry, and I would also make a reference to the degree of African ancestry throughout the different Latin American groups both between nations an within nations. It would be a report on actual ancestry, and not misleading report such as those of newspapers which group categories from the census and form new ones to arrive at quite big misrepresentations of a country, such as they usually do with Brazil.

There is nothing wrong making explicit how the ~100 million figure was arrived at, and besides I have posted a link with the official information by the Brazilian government. On the same article, the genetic studies, at least when it comes to Brazil, show quite clearly that African ancestry is widespread among Brazilians (no matter if 'white', 'pardo' or 'black') and it also shows the genetic estimates of African input in the Brazilian population as a whole, the degree of African contribution.

I think that's quite a fair representation, no one here is downplaying it, but no exaggeration is being made either. Besides, after reading the articles on pardo Brazilians, the genetic studies, etc, one ca make his own conclusions about what Brazil is like.Grenzer22 (talk) 11:24, 23 January 2014 (UTC)

I have never suggested we use the Guardian article as a source, nor any source that says half of Brazil are "like Nigeria" (a stupid statement when you consider that Nigeria has over 250 different ethnic and cultural groups speaking some 500+ languages - "like Nigeria" can have no real meaning in such a diverse context). The source originally used in the article[13] is a report from a 10-day fact-finding mission in Brazil by the UN panel on Afro-descendants which arrived at it's own estimate of African ancestry without reference to the census categories at all. Other reliable sources such as the Minority Rights Group are also completely independent of the census, and the statement by the Brazilian state prosecutor makes no mention of the census either. If you read the sources, the articles which use the census figures generally talk about "black and mixed-race" people, while independent ancestry estimates use terms like "Afro-Brazilians" or "with African ancestry" instead and this is the figure we want. It's a little bit confusing because the final figures are approximately the same, but the 100M estimate for "African ancestry" has nothing to do with combining the "black" and "pardo" figures on the census.

The Venezuelan government source gives a direct estimation of African ancestry ("Afro-Descendent"), and I note that it's completely different to what we'd get if we just added the "black" + "moreno" figures from the census. If you can find an official Brazilian government source that explicitly states how many people have African ancestry (as opposed to how many belong in various colour groupings) then I'd be happy to use that instead.

Please understand that I'm not trying to misrepresent Brazil, I'm just trying to make a pop box that is informative and uncluttered, and consistent across all countries. The points you raise about "black" and "pardo" are very important and should be discussed, but the pop box should just list a figure representing "African ancestry". The details and explanations belong in the Brazil section, just like with all the other countries.

Tobus (talk) 14:10, 23 January 2014 (UTC)

This quote (http://www.ctvnews.ca/world/racism-remains-institutionalized-in-brazil-un-panel-1.1591369#ixzz2qiHVgKK9) does not give the 100 million figure, but the other quotes posted above (like the Guardian article and the other sources that you have provided in the article) do (that's the 'reasoning' behind the 100 million figure, nothing else; 43% pardo + 7% black Brazilians; you can check it here, the Reuters article on the prosecutor mentions it: "In the Reuters interview, Santos also added that the host should have been representative of Brazil’s racial makeup. He said nearly half of the Brazilian population identify with being Afro-Brazilian. In fact, the 2010 Brazil Census found that the majority of the population declared themselves as black or brown, http://fusion.net/culture/story/world-cup-event-called-racist-white-host-296177; but again, they did not mention that brown Brazilians would be 43%, and 7% 'black', rather they again placed 'black' ahead of 'pardo' as if their % were higher; and so does your minority groups report: "They number 80 million or 48 per cent of the total population in official statistics [...] Afro-Brazilians are categorized in the census as mixed race, pardo or preto"; a clear reference to the census figure, and without really explaining it, since many people of substantial Native American instead of African, as in Northern Brazil, besides others who are quite pred. non African in ancestry, etc, may also report as pardo to the census; http://www.refworld.org/docid/49749d4d32.html).

The UN worker simply said that over half of Brazilians would have African ancestry (http://www.ctvnews.ca/world/racism-remains-institutionalized-in-brazil-un-panel-1.1591369#ixzz2qiHVgKK9), which is true. But in fact over half of Venezuelans, over half of Colombians, etc, in short, the average Latin American, also has African ancestry. So we're not going here by any degree of African ancestry, but rather and in fact by random and various sources. And if it is so, the census should be used, and usually some people lump black and pardo to arrive at the figure of 100 million, though this is highly controversial, but anyway, since it is the census information, it should be kept, with the explanation.

The UN worker did not conduct any measurement of African ancestry in Brazil, and the census anyway is undertaken only by the Brazilian government. Therefore the information should be kept as it is now.Grenzer22 (talk) 15:49, 23 January 2014 (UTC)

Another article you quoted Tobus:

"About half of Brazil's 204 million people are black — more than in any nation except Nigeria".http://www.huffingtonpost.com/2013/11/06/brazil-government-job-blacks_n_4224948.html How do you think they came up with the figure? By adding the black and pardo Brazilians (7%+43% or roughly 50% of the Brazilian population).

They don't explain that in fact white, pardo and black Brazilians, ancestrally, are actually mixed. They label pardo and black Brazilians as the same, they ignore the (overwhelming) non african ancestry of pardo ('mixed race' Brazilians), and they fail to mention that pardo Brazilians are 43% of the population and black Brazilians are 7%. Actually, they take both as 'blacks', as if they were all (including pardo Brazilians) similar to African Americans or Nigerians, who are pred. African in ancestry, whereas pardo Brazilians, who comprise 43% of the 50% figure are in fact for the most part quite pred. non african (reaching as high as 90% non african ancestry in Northern Brazil).

I think for the sake of accuracy one should point out the "reasoning" behind the figure. And the "reasoning" as I just showed to you on the the Guardian article above is lumping black and pardo Brazilians (~43% + ~7% = ~50%). Often, as I showed, accompanied by quite a few misleading conclusions (such as that half of Brazil would be like Nigeria - as in fully or near fully African like Nigerians, etc). I haven't changed the figure. I pointed out how it came about (it does not come out of measuring African ancestry in Brazilians, or out of a census conducted on their own, but simply out of lumping black and pardo figures from the Brazilians census). As I said, if the article were really about any degree of African ancestry, we would have to raise the figure for all other Latin American countries.

In short, the census figure should be kept.Grenzer22 (talk) 16:06, 23 January 2014 (UTC)

As I said, there is nothing wrong making explicit how the ~100 million figure was arrived at, and besides I have posted a link with the official information by the Brazilian government. On the same article, the genetic studies, at least when it comes to Brazil, show quite clearly that African ancestry is widespread among Brazilians (no matter if 'white', 'pardo' or 'black') and it also shows the genetic estimates of African input in the Brazilian population as a whole, the degree of African contribution.I think that's quite a fair representation, no one here is downplaying it, but no exaggeration is being made either. Besides, after reading the articles on pardo Brazilians, the genetic studies, etc, one can make his own conclusions about what Brazil is like.Grenzer22 (talk) 16:16, 23 January 2014 (UTC)

If you do the maths it's clear there are two separate figures being talked about: "African ancestry" and "black + pardo". Santos[14] says "nearly half" (ie less than 50%) have African ancestry but "the majority" (ie more than 50%) are black or brown - clearly two different figures. Minority Rights give an official figure of 80M people with African ancestry, but if this was added up from them being "categorized in the census as mixed race, pardo or preto" it would give a figure of 100M - clearly the 80M "African ancestry" is a different figure to the "black + pardo" figure from the census. The UN panel[15] would have had access to a wide range on internal documents as well as their own research and they make no mention of the census or use it's categories. The "over half" assessment is an independent estimation.

The other sources are all using the "black + brown" from the census as you say, but either don't, or mistakenly, report it as "African ancestry". The Guardian[16] says "African brazilians" in the headline but uses "black or mixed race" when refering to the actual data, the same report at the more reliable BBC[17] uses "non-white" throughout and never states the figure as being African ancestry - the Guardian is wrongly (deliberately?) assuming "black + brown" = "African ancestry" to get an attention grabbing headline. The articles which talk about Nigeria are also using the census but never claim "African ancestry"... they are the least reliable sources we have so we wouldn't use them in any case.

So the 100M you are using might come from adding "black+brown" from the census, but the 100M I'm using doesn't. In the previous discussion I originally suggested using 80M - the only official estimate we have, but was persuaded by another user to use the 100M figure as this seems to be a the general consensus in the media and is supported by the UN panel. That user has since been banned as a WP:SOCK so I feel no obligation to honour that discussion. To resolve this issue and remove any confusion about the figure being "black + brown" from the census, I suggest we use the official figure of 80M - it's probably a bit low but it is the only official government estimation we have, and it is clearly a direct estimate of "African ancestry", not a colour combination.

Tobus (talk) 01:06, 24 January 2014 (UTC)

Tobus, Santos also lumped pardo and black Brazilians (nearly is ~50%), and so did the minority groups report in the quote I gave. There are no "internal documents" or any genetic measurement which would back up the ~50% figure but the census, and lumping in black and pardo Brazilians, even if this is controversial. If you would see no problem we could make it ~100 million (7% black + if including 43% pardo Brazilians). We know that is not the correct figure, but as I have shown, and you can believe me on it, the sources usually do it, they add black and pardo Brazilians. There has not been any independent census by the UN or any genetic measurement on this by the UN or sources, they just indirectly used the census. As a Brazilian, I'm used to seeing it.

As I said, if it is about any degree of African ancestry, the figure would have to be raised not only to Brazil, since the vast majority of Brazilians ('white', 'pardo' and 'blacks') have African ancestry, and this is shown in the genetic studies in the article, but also in other Latin American countries the average person does have at least some degree of African ancestry. Personally, I'd prefer this approach, since it would be the most truthful, but I'm afraid it would raise too much controversy, even if the genetic studies are pretty clear on it. I'd add the degree of African ancestry both between and within nations, but many from other Latin American countries probably would not allow it, I'm glad it has been done at least with Brazil (and it does show the widespread African ancestry in Brazilians).

Look at what genetic studies show for Venezuela, f.e:

BACKGROUND: The Venezuelan population is the product of Native American, African and European admixture. Few admixture studies have been made in Venezuela using short tandem repeats (STRs).

RESULTS: For the Northern-Central Region, the estimate of admixture was 37.7% for the European component, 37.7% for the African and 24.6% for the Native American. For the Central-Western region, the estimate of admixture was 58.5% for the European, 16.5% for the African and 25.0% for the Native American component. http://www.ncbi.nlm.nih.gov/pubmed/17536755

As one can see the average Venezuelan does have African ancestry, and the figure on the text is clearly downplaying it (much more than with Brazil, since the figure when it comes to Brazil comes out adding self reported 'black' and 'pardo' figures, whereas the Venezuelan does not. The same goes for Colombia and the other Latin American countries.Grenzer22 (talk) 02:10, 24 January 2014 (UTC)

If you don't agree with my proposal ~100 million (7% black + if including 43% pardo Brazilians; plus the links I gave), I'd rather prefer you to keep the ~100 million vs the ~80 million since it is closer to the truth, and preferably keeping the IBGE and the genetic study covering the whole of Brazil I provided in the box. Best regards.Grenzer22 (talk) 02:14, 24 January 2014 (UTC)

I agree that the 100M is probably more accurate and best reflects the general consensus on the issue. As I've already stated wikipedia policy doesn't allow WP:SYNTH on our part so we need a source that either states "African ancestry" directly, or does the "black+pardo" synthesis itself - neither the census nor the genetic study you cited do this. My original choice was the UN panel as it seems the most independent and reliable, but I am willing to use a any source that meets the WP:Reliable Source policy. As a suggestion, perhaps we can use the Santos source instead - it's from a Brazilian government employee and the article states both interpretations, "Afro-brazilian" and "black+mixed-race". Would you accept this: "Brazil: 100M[18]"? Tobus (talk) 23:33, 24 January 2014 (UTC)

^ Like the Guardian article and the minority groups report, the prosecutor is lumping pardo and black Brazilians, though as I said this is controversial. Anyway, the census link, besides being official information, fits right in with it, and this way has been employed here and elsewhere without opposition. If the person wants to read further, he will move down the section on this article, and check that African ancestry is quite widespread, and move on to the pardo article too. This is how conventionally the figure is arrived at. The prosecutor did not make any measurement of African ancestry nor conducted a census, he just used the census figures, as the other articles quote above (minority groups report, the Guardian, etc).

One could just add a ~100 million, and keep the clarification with the official census. In the Brazilian section of the article I think it is fairly well explained that African ancestry is quite spread according to genetic measurements of actual ancestry. The census is based on self reported ancestry, whereas the genetic studies aim at actual ancestry. So this way it would be the most informative.Grenzer22 (talk) 02:10, 25 January 2014 (UTC)

The pop box is called "Estimated Ancestry with African Ancestry" and this figure for Brazil is 100M, however it is arrived at. There is no need to elaborate on this inside the pop box itself (and no space to do it properly), and it's not done for any other country even though most of them don't usually count Afro-descendants as a single group either. As per the title and the format of the other populations, "100M" is all that should go in this particular pop box.

The source for this figure has to be one that states it directly, we can't (as per WP:OR) calculate it ourselves, nor (as per WP:SYNTH) can we combine two sources to arrive at a conclusion that neither of them explicitly state. We can however cite some other WP:RS which does calculate it themselves and states it as fact. We can't use the census (which doesn't report "African ancestry" directly), nor can we use the census plus a genetic study (neither of which explicitly state the conclusion). That leaves us with one of the other sources that does say "half the population" has African ancestry. I don't mind which one, I'd prefer the UN, but Santos mentions the "black/mixed-race" combination as well so you might prefer that. Any of the others would also be OK, so long as it explicitly states "African ancestry" or "Afro-Brazilian".

My apologies if it seems I'm being picky, but the pop box gets a lot of attention from people trying to push their POV and so the text and the source need to be watertight. Whatever we agree on I don't want somebody to be able to change it because we synthed the figure or failed to source it properly. Both of the sources you've provided are high quality and are used correctly elsewhere in the page, but they don't state the "100M African ancestry" figure directly so we can't use them for the pop box.

It's largely irrelevant since we've agreed not to use it, but I'll point out that the Minority Rights report came out in 2008 and the census came out in 2010 - there's no way they could have used it. The previous census in 2000 counted "black" 6.8% and "pardo" 38.5, totalling 44.7% so the report's 48% didn't come from that census either - the Minority Rights Group figure definitely comes from somewhere other than combining "black" and "pardo" on the census.

Tobus (talk) 07:55, 25 January 2014 (UTC)

The minority groups report does mention how they arrived at it: "They number 80 million or 48 per cent of the total population in official statistics [...] Afro-Brazilians are categorized in the census as mixed race, pardo or preto"" http://www.refworld.org/docid/49749d4d32.html). They say it quite clearly that the 'Afro Brazilians' would be the 'mixed race pardo' and the 'black' Brazilians. The 48% figure is right in there with the various census estimates of lumping pardo and black Brazilians (there were census estimates by the IBGE between 2000 and 2010, and they fit right in with the 44% 2000 figure and the 50% 2010 figure). As I said, they did not conduct any measurement of African ancestry, they did not carry out any census, they just did what some people usually do, based on the census, which is highly controversial as I pointed out. None of the sources mention any independent measurement of African ancestry, nor an independent census, simply because they did not exist. The key figure is provided by the Brazilian census, as reported by Brazilian themselves and this should be respected.

The Brazilian figure is an important figure, and it is important to explain how it was arrived at, and the census link is the key information, since the various sources which give the ~100M figure do lump pardo and black Brazilians. This explanation is important, and whether or not it is made in reference to other countries, it should be there on the Brazilian population, since it clarifies the subject.

Anyway, the census link, besides being official information, fits right in with it, and this way has been employed here and elsewhere without opposition. If the person wants to read further, he will move down the section on this article, and check that African ancestry is quite widespread, and move on to the pardo article too. This is how conventionally the figure is arrived at, and it should be clarified (as I said, it has been done in other articles, and no opposition was raised, I don't see any problem with it).Grenzer22 (talk) 12:51, 25 January 2014 (UTC)

I think you have misunderstood the Minority Rights Group report, but as I said, it's not relevant so I won't continue that part of the discussion.

I agree the Brazilian figure is an important figure and agree that it is important to explain how it was arrived at - but I don't think it's important to do it in this particular pop box. The explanation should be left to the Brazil section like all the other countries.

The census itself does not state a figure for the number of individuals with African ancestry - it relies on WP:SYNTH on your part to reach the figure, and that synthesis is demonstrably incorrect according to the genetic studies ("The high ancestral variability observed in Whites and Blacks suggests that each Brazilian has a singular and quite individual proportion of European, African and Amerindian ancestry in his/her mosaic genomes. Thus, the only possible basis to deal with genetic variation in Brazilians is not by considering them as members of colour groups, but on a person-by-person basis".[19] The strange thing about this about this situation is that other reliable sources have made that synthesis and it has become a widespread "fact", so we are entitled (in fact, obligated) to report it as such. We still can't use the census as the source though because it doesn't make the statement itself. Do you see what I'm getting at?

If the census has been used as a source for "African ancestry" elsewhere that doesn't mean we should use it here. If you point me to the places you think it's been used like this I'll take a look and explain why it's OK in that context or if it's an incorrect usage.

Tobus (talk) 01:31, 26 January 2014 (UTC)

I agree with the geneticist explanation you have quoted: "Thus, the only possible basis to deal with genetic variation in Brazilians is not by considering them as members of colour groups, but on a person-by-person basis". (http://www.scielo.br/scielo.php?pid=S0100-879X2009005000026&script=sci_arttext#Abstract) This is my opinion too, and the reality.

Since the wikipedia does have these categories, and they are routinely played out by both governments and the press, I was just trying to clarify how the 100 million figure was arrived at, nearly the majority of the texts you mentioned make a reference to it, and even the one which does not (and it is also the one which does not mention the 100 million figure, that one by the UN) implicity is making a reference to it, since they have not carried out an independent census nor genetic measurement. The African American population article African American, f.e, in its box clearly states the number of 'mixed race' African Americans, giving a number, even if we know that African Americans as a group, not a tiny minority, are 'mixed'. I am not relying on WP:SYNTH to explain how the figure was reached: from the articles you posted I showed quite clearly that's what they routinely do, and even if it were not so, one could easily find other articles showing that the ~50% is arrived at by lumping pardo and black Brazilians.

I'm not trying to argue with you, and I wouldn't do it as a sport. Try to understand me, Brazil is poorly explained not only by the Brazilian government but especially by sources abroad. The 100M figure is misleading and many will and already do misinterpret it(as I showed to you based on that American report).

I think having a link showing that the sources lump pardo and black as a single category to explain how the figure was arrived at is just a necessary clarification. Even if that kind of lumping is highly controversial and not really a mirror of reality. But it is better IMO.

As I said, if it is about any degree of African ancestry, the majority of Latin Americans do have African ancestry, and the figure would have to be raised to ~500 million. This way, I think the article is misleading too, by giving the wrong impression that the % of Brazilians who have any degree of African would be higher than in other Latin Americans, when it is not: it is the degree of African ancestry in Brazilians which is usually higher than that of other Latin Americans, not the % of those who have any degree of African ancestry (all genetic studies show this very clearly).

Anyway, since this is a free encyclopedia and its scope is to inform, the more clarified the subject is, the better it will be. Best regards.Grenzer22 (talk) 12:17, 26 January 2014 (UTC)

Tobus, you clearly are a reasonable person, and I was just wanted to make clear the nuances of Brazil which are often overlooked abroad, and sometimes even in Brazil. If you want to revert back to that edition of yours, I will not oppose it. If you want to leave it as it is now, it'll be fine too (and better IMO). Please just leave the explanation below in the article, which I think is a reasonable one.Grenzer22 (talk) 12:35, 26 January 2014 (UTC)

The African-American info box is describing a single population group and I think it makes sense to have that kind of information there - the same with the info box on the Afro-Brazilian (which I just noticed is missing the "pardo" population!). The pop box on this article though is different - it's giving the populations for a large number of different countries from an outside perspective and it would be confusing and messy to try to include details of how each country interprets the topic. For that reason I will go back to the single "100M" figure.

I will leave the explanation in the main text as it is - as I said before I think what you want to explain is very imporant and should definitely be included in the article, it's just out of place in this particular pop box. One thing I'll ask you to reconsider though is the recent edit to add "if including the pardo Brazilian population" in the History section - the section is a global overview and the sentence already states "with significant black, mulatto, or zambo populations" so it's already saying "pardo", just using more general terms. If you think the Brazilian-specific term "pardo" is important to mention at that particular point then perhaps it can be incorporated into the lead-in phrase instead (eg "with significant black and mixed-race pardo, mulatto and zambo) populations").. or perhaps just saying "mixed-race" is enough since that's "pardo" in English? I will leave that at your discretion.

Tobus (talk) 02:30, 27 January 2014 (UTC)

The pardo reference in the history section is a necessary explanation, and it includes the category as it is named in Brazil. It is not in the box, so there should be no problems.Grenzer22 (talk) 11:16, 27 January 2014 (UTC)

Changes to Image Box need Talk Page

Latest comment: 10 years ago2 comments2 people in discussion

I am just jumping in because I deal with a lot of these image box issues. The best way to make ANY changes to an image box (which is a popular source of disagreement) is via the talk page. I have always added a hidden tag that NO major alterations should be made without first discussing it on the talk page. This makes sense to maintain some order and prevent unilateral changes.--Inayity (talk) 05:45, 26 April 2014 (UTC)

Normally, I would totally agree. However, being that mixed ancestry is not sourced with being coined the term Afro-Latin American, it did not seem fitting to me and thus not the logical step to take. If there was minimal citation, or sources that were not as reputable as they should be, I would have opened up a discussion first. There has been constant reverts without any effort to improve the article surrounding the issue at hand with the questionability between mixed people and those who are of mostly of African ancestry. According to the CIA consensus these account for separate things. The consensus on the talk page that I am getting, is that the picture box should reflect the Afro diversity including mixes (even minimal Afro ancestry), which is not a good argument when the term itself is in question for sourcing. But in most circumstances, I totally agree that a discussion should be made first. If you can contribute in any way to this, it would be greatly appreciated. Thank you kindly. Savvyjack23 (talk) 07:12, 26 April 2014 (UTC)

^ http://www.plosone.org/article/fetchObject.action?uri=info%3Adoi%2F10.1371%2Fjournal.pone.0075145&representation=PDF
^ http://cienciahoje.uol.com.br/noticias/2011/02/nossa-heranca-europeia/?searchterm=Pena
^ Cite error: The named reference plosone.org was invoked but never defined (see the help page).
^ Cite error: The named reference onlinelibrary.wiley.com was invoked but never defined (see the help page).
^ Forensic Science International: Genetics. Allele frequencies of 15 STRs in a representative sample of the Brazilian population (inglés) basandos en estudios del IBGE de 2008. Se presentaron muestras de 12.886 individuos de distintas etnias, por regiones, provenían en un 8,26% del Norte, 23,86% del Nordeste, 4,79% del Centro-Oeste, 10,32% del Sudeste y 52,77% del Sur.
^ ^a ^b ^c ^d Cite error: The named reference pmid21359226 was invoked but never defined (see the help page).

[1] ttp://www.plosone.org/article/fetchObject.action?uri=info%3Adoi%2F10.1371%2Fjournal.pone.0075145&representation=PDF

[2] ttp://cienciahoje.uol.com.br/noticias/2011/02/nossa-heranca-europeia/?searchterm=Pena

[plosone.org-3] Cite error: The named reference plosone.org was invoked but never defined (see the help page).

[onlinelibrary.wiley.com-4] Cite error: The named reference onlinelibrary.wiley.com was invoked but never defined (see the help page).

[5] Forensic Science International: Genetics. Allele frequencies of 15 STRs in a representative sample of the Brazilian population (inglés) basandos en estudios del IBGE de 2008. Se presentaron muestras de 12.886 individuos de distintas etnias, por regiones, provenían en un 8,26% del Norte, 23,86% del Nordeste, 4,79% del Centro-Oeste, 10,32% del Sudeste y 52,77% del Sur.

[pmid21359226-6] Cite error: The named reference pmid21359226 was invoked but never defined (see the help page).

[1]

[2]

[3]

[4]

[5]

[6]