Scientists Decry "Flawed" and "Horrifying" Nationality Tests(blogs.sciencemag.org)
blogs.sciencemag.org
Scientists Decry "Flawed" and "Horrifying" Nationality Tests
http://blogs.sciencemag.org/scienceinsider/2009/09/border-agencys.html
5 comments
Linguistics would offer a near unparalleled way to tell. Linguistic specialists are used by the police in the UK to locate where hostage takers (or ransom makers) come from.
IIRC there's 3 categories they can tell on average; where you grew up, where you lived and where you've lived recently. Basically if you grew up in London, lived in Manchester and have lived in Scotland for a half-year or so, they'll know. When compared to police records of hardened criminals this can usually help them narrow it down to a handful of people, sometimes when a persons accents are so strong (IE you can locate a specific area of each city) it can basically get you the person you're dealing with. It then gives negotiators the upper hand, because they know who they're dealing with and their behaviours from previous crimes before they've even opened negotiations.
If linguistic specialists can work this well for accents inside the UK, I'm sure they'll be able to tell if you're lying when you say you grew up in Somalia to Somali parents, but you're accents indicating you're from Kenya or Ethiopia. It might not be useful all the time, but it could certainly help add credence to someone's story and potentially help a refugee with legitimate claim through faster.
I doubt you'd be able to narrow down regionality through linguistics like you can in the UK. British accents are usually so strong that they're like prime colours in language, I've got friends here in Canada who immigrated here when they were infants, but they've still got a British accent in there. When you're dealing with adults, the accent is as clear as day even when they've been in Canada for decades.
IIRC there's 3 categories they can tell on average; where you grew up, where you lived and where you've lived recently. Basically if you grew up in London, lived in Manchester and have lived in Scotland for a half-year or so, they'll know. When compared to police records of hardened criminals this can usually help them narrow it down to a handful of people, sometimes when a persons accents are so strong (IE you can locate a specific area of each city) it can basically get you the person you're dealing with. It then gives negotiators the upper hand, because they know who they're dealing with and their behaviours from previous crimes before they've even opened negotiations.
If linguistic specialists can work this well for accents inside the UK, I'm sure they'll be able to tell if you're lying when you say you grew up in Somalia to Somali parents, but you're accents indicating you're from Kenya or Ethiopia. It might not be useful all the time, but it could certainly help add credence to someone's story and potentially help a refugee with legitimate claim through faster.
I doubt you'd be able to narrow down regionality through linguistics like you can in the UK. British accents are usually so strong that they're like prime colours in language, I've got friends here in Canada who immigrated here when they were infants, but they've still got a British accent in there. When you're dealing with adults, the accent is as clear as day even when they've been in Canada for decades.
If linguistic specialists can work this well for accents inside the UK, I'm sure they'll be able to tell if you're lying when you say you grew up in Somalia to Somali parents, but you're accents indicating you're from Kenya or Ethiopia.
The boundaries between Somalia, Kenya, and Ethiopia were drawn by European diplomats in the nineteenth century for their convenience. Those boundaries cut across the territory occupied the ethnically-Somali people themselves, who (if Wikipedia can be trusted) are traditionally nomadic and who frequently marry outside their own clans. So if some linguist claimed that he could identify the origin of a Somali-speaker the way Henry Higgins could trace an English-speaker, I would be very very skeptical.
The boundaries between Somalia, Kenya, and Ethiopia were drawn by European diplomats in the nineteenth century for their convenience. Those boundaries cut across the territory occupied the ethnically-Somali people themselves, who (if Wikipedia can be trusted) are traditionally nomadic and who frequently marry outside their own clans. So if some linguist claimed that he could identify the origin of a Somali-speaker the way Henry Higgins could trace an English-speaker, I would be very very skeptical.
I could :-) The borders between Somalia and Kenya is one thing, with Ethiopia another.
The Kenyan border has always been a no-go zone for most ethnic Somalis. That area has the highest concentration of Somali minorities, both ethnic and linguistic. The mainstream Somalis went there after recent famines and some where relocated there by the government in the 60s and 70s. Previously they have gone there back and forth in pastoral voyages, but always "returned", or more accurately, they found other places where the grass was, literally, greener.
Somali Kenyans (Somali-Sujuu, not recent refugee) are Darood from the North and North West, and not other Somali tribes from the middle and the south of the country, closer to Kenya :-) The Sujuu concentrate in Garrise, Mandheere and Wajeer cities in Kenya. They have funny Somali accents, but they're considered proper Somalis, since they are mostly Ogaden tribe who are known for bravery (the recipients of the first air-strike outside of Europe. They also hold the dubious title of being at war with every colonizing force in Somalia, simultaneously; Ethiopian Emperor, Great Britain, Italy and France.)
The Ethiopian border with Somali IS, however, arbitrarily drawn. One fifth of Ethiopia's population are ethnic Somalis.
However, Djibouti is one country that has a Somali population of 60% and they're happy not to call themselves Somalis.
The Kenyan border has always been a no-go zone for most ethnic Somalis. That area has the highest concentration of Somali minorities, both ethnic and linguistic. The mainstream Somalis went there after recent famines and some where relocated there by the government in the 60s and 70s. Previously they have gone there back and forth in pastoral voyages, but always "returned", or more accurately, they found other places where the grass was, literally, greener.
Somali Kenyans (Somali-Sujuu, not recent refugee) are Darood from the North and North West, and not other Somali tribes from the middle and the south of the country, closer to Kenya :-) The Sujuu concentrate in Garrise, Mandheere and Wajeer cities in Kenya. They have funny Somali accents, but they're considered proper Somalis, since they are mostly Ogaden tribe who are known for bravery (the recipients of the first air-strike outside of Europe. They also hold the dubious title of being at war with every colonizing force in Somalia, simultaneously; Ethiopian Emperor, Great Britain, Italy and France.)
The Ethiopian border with Somali IS, however, arbitrarily drawn. One fifth of Ethiopia's population are ethnic Somalis.
However, Djibouti is one country that has a Somali population of 60% and they're happy not to call themselves Somalis.
IIRC there's 3 categories they can tell on average; where you grew up, where you lived and where you've lived recently. Basically if you grew up in London, lived in Manchester and have lived in Scotland for a half-year or so, they'll know.
If linguistic specialists can work this well....
Law enforcement have a bad track record of having great faith in techniques that don't actually work. One recent example in the news was the bogus forensic arson investigation techniques that were used to convict and execute Cameron Todd Willingham. Confessions are another example -- how many cases have been closed prematurely because police believed it was implausible for a person with no obvious mental abnormalities to confess to a heinous crime they didn't commit? That's why it's important to scientifically validate the techniques that are used. The fact that police have used a method and found it completely satisfactory doesn't say much about its validity.
If linguistic specialists can work this well....
Law enforcement have a bad track record of having great faith in techniques that don't actually work. One recent example in the news was the bogus forensic arson investigation techniques that were used to convict and execute Cameron Todd Willingham. Confessions are another example -- how many cases have been closed prematurely because police believed it was implausible for a person with no obvious mental abnormalities to confess to a heinous crime they didn't commit? That's why it's important to scientifically validate the techniques that are used. The fact that police have used a method and found it completely satisfactory doesn't say much about its validity.
Law enforcement have a bad track record of having great faith in techniques that don't actually work.
Yes, actually most police investigation techniques have never been subjected to proper validation studies.
Yes, actually most police investigation techniques have never been subjected to proper validation studies.
I am Somali and it seems like I might not be able to pass this test along with every Somali-Arab. Neither will Marka people (with mixed Arab, Portuguese and Persian stocks) or the various Bantu groups in the south, or the small community of Italian misiyooni, or Zaylici people with Turkish lineage, or the Punjabis in Kismayo, or the Ashraf tribes who came from Saudi Arabia 200 years ago.
However, Egyptians, Libyans and Berber north Africans will pass with flying colors.
http://en.wikipedia.org/wiki/Somali_people#mtDNA
Just in my own family, we have people who are half Italian, Emirati, and Thai.
However, Somalis have a fairly secure system for identifying one's lineage. We know our fathers' names (ignoring infidelity here) for up to 20 names or more; I can do 39. And then there is the extensive clan system where anyone is able to name their tribe down to the minutest family name. Ask them "waar abtirso!", i.e. name your lineage, and any Somali should be able to do so. The language itself is a sufficient enough cipher; non-Somalis, at least those from that part of Africa will not be able to pronounce certain sounds, like Dh, C, X, Kh, and Q.
I am not "visibly" Somali and most Somalis are awed when I speak to them, however, it just takes a brief moment of mutual introduction then it's all hugs and kisses.
This is just a case of a private commercial interest pushing an expensive non-solution to the government. Instead of the cheaper alternative: higher Somali-British immigration officers, if you're afraid of corruption, higher N officers and let them perform the vetting process in a secret-vote manner, the officers are ignorant of each other and it takes (N/2)+1 votes to admit someone.
However, Egyptians, Libyans and Berber north Africans will pass with flying colors.
http://en.wikipedia.org/wiki/Somali_people#mtDNA
Just in my own family, we have people who are half Italian, Emirati, and Thai.
However, Somalis have a fairly secure system for identifying one's lineage. We know our fathers' names (ignoring infidelity here) for up to 20 names or more; I can do 39. And then there is the extensive clan system where anyone is able to name their tribe down to the minutest family name. Ask them "waar abtirso!", i.e. name your lineage, and any Somali should be able to do so. The language itself is a sufficient enough cipher; non-Somalis, at least those from that part of Africa will not be able to pronounce certain sounds, like Dh, C, X, Kh, and Q.
I am not "visibly" Somali and most Somalis are awed when I speak to them, however, it just takes a brief moment of mutual introduction then it's all hugs and kisses.
This is just a case of a private commercial interest pushing an expensive non-solution to the government. Instead of the cheaper alternative: higher Somali-British immigration officers, if you're afraid of corruption, higher N officers and let them perform the vetting process in a secret-vote manner, the officers are ignorant of each other and it takes (N/2)+1 votes to admit someone.
I think the wording is a bit strong- it's really not that horrifying- but it's certainly not scientifically sound. It's not like there's a mountain range surrounding Somalia that has prevented them from interbreeding with individuals from other countries for the last hundred thousand years. Heck, the place didn't really even exist until the 1950s.
Well it is horrifying, they're planning on refusing people who are immigrating on humanitarian and compassionate grounds (refugee) based on their DNA.
Warfare doesn't check your DNA before it kills your spouse and children. Apparently Britain, however, checks your DNA before it decides to save you or not.
Warfare doesn't check your DNA before it kills your spouse and children. Apparently Britain, however, checks your DNA before it decides to save you or not.
So you believe in no immigration restrictions at all then? I think it's perfectly fair to only admit people into a country if they desperately need it. They just want a way to identify those people. This is the wrong way to do it, but that's a separate issue.
True, but the real villains here are the non-refugee immigrants who lie about their country of origin in order to get refugee status.
It's also horrifying because of how the government is using utterly misguided "science". It's like if they demanded everyone get homeopathic therapy.
It's not quite as bad as homeopathy. DNA evidence is worth something in these cases, but it's not conclusive -- the main problem is that government bureaucrats are too dumb to understand this and would treat it as conclusive.
If we knew the number of Somali citizens with "non-Somali" DNA and the number of non-Somali citizens with "Somali" DNA, then... well, there's another application for Bayesian statistics, I guess.
If we knew the number of Somali citizens with "non-Somali" DNA and the number of non-Somali citizens with "Somali" DNA, then... well, there's another application for Bayesian statistics, I guess.
Thank god I'm emigrating not immigrating to the UK.
First 'Citizenship Education', I said it then when I refused to take the course and now; you've got to be fucking stupid. Last time British people were nationalistic we committed genocide across the Americas, Africa and Australia, with a few places here and there on the way.
Now the governments gone about ten steps dumber and now believe mitochondrial and Y chromosomal DNA corresponds to nationality. So when my Canadian immigration papers come through, where do I go to get my DNA changed? Oh wait, everyone here's still fucking British! (Aside from Quebec and NB, but you're likely to pick up as many French Chromosomes in the rest of Canada as you would in the UK)
First 'Citizenship Education', I said it then when I refused to take the course and now; you've got to be fucking stupid. Last time British people were nationalistic we committed genocide across the Americas, Africa and Australia, with a few places here and there on the way.
Now the governments gone about ten steps dumber and now believe mitochondrial and Y chromosomal DNA corresponds to nationality. So when my Canadian immigration papers come through, where do I go to get my DNA changed? Oh wait, everyone here's still fucking British! (Aside from Quebec and NB, but you're likely to pick up as many French Chromosomes in the rest of Canada as you would in the UK)
Should be extended to all British citizens. Then those without British DNA can be kicked out.
That will get rid of a whole lot of useless <beeps> in the house of Lords with a 'de', 'le' or 'ville' in their names.
And it will allow a large number of new people in. After a few hundred years of determined DNA distribution through the empire there must be quite a few Indians, Africans and even Americans who have some English/Scottish or Irish in them (or at least their ancestors did !)
That will get rid of a whole lot of useless <beeps> in the house of Lords with a 'de', 'le' or 'ville' in their names.
And it will allow a large number of new people in. After a few hundred years of determined DNA distribution through the empire there must be quite a few Indians, Africans and even Americans who have some English/Scottish or Irish in them (or at least their ancestors did !)
This is a good response to the commercial services that claim to be able to place people's ancestors to very specific locations in continents that have been very poorly surveyed for geographically correlated genome variation. (The worldwide number of well located, well analyzed human genome variant samples is still shockingly small.) Other forms of investigation would be much more informative of an asylum-seeker's genuine national origin.