# Distinguishing half-brother from uncle when sharing 2,183 centimorgans?

Results from Ancestry show that I share 2183 centimorgans with a complete stranger who is my age.

• 1450-2050 centimorgans = grandpa, uncle, half-brother
• 2400-2800 centimorgans = brother

I wonder if the fact that we share so many centimorgans (somewhere between the two above categories) makes it less likely that he is my uncle and more likely that he is my half-brother. Or how should I interpret an amount of centimorgans which is between the two ranges?

If I understand correctly, the figure below leaves little room for interpretation that we are half-siblings — but I’m no expert and don’t want to get too far ahead of myself.

The Shared cM Project 3.0 tool v4 suggests for 2,183 that the most likely relationships are:

• grandparent/grandchild; or
• half sibling

and uncle/nephew falls just outside the expected range with a comment of:

this relationship has a positive probability for 2183cM in thednageek's table of probabilities, but falls outside the bounds of the recorded cM range (99th percentile)

Based on that comment and your ages the most likely relationship seems to be half sibling.

Your stated numbers are off. I don't want to edit it without your okay.

Half sibling is 1317-2312 cM, so your match of 2183 cM fits well. You say the range ends at 2050 (you're combining several ranges into one).

Sibling is 2209-3384 cM, which is different from what you claim. Your match is just a bit too small to fit there, but it could be an outlier. Half sibling is more likely though.

How can you tell the difference between full and half sibling? Get both sets of raw data into a place with a chromosome browser, such as Gedmatch.com. Full siblings will have some segments with full matching (both sides). Half siblings will not.

Uncle is 1349-2175 cM. Not impossible to be an outlier, but not very likely.

Grandparent is 1156-2311 cM. Your match fits well here. But you can rule it out based on your ages being similar. You'd need at least 30 years of separation to be grandparent/grandchild (rare cases might be less but never below 20 years and that would be insanely rare, even 30 is very uncommon).

Half sibling is your most likely match. Another possibility is that your lineage includes some close relative relationships that are throwing off the numbers. But let's assume standard relationships for now.

In her webinar DNA Rights and Wrongs: The Ethical Side of Testing, Judy G. Russell, JD, CG, CGL cited an incident that happened after a keynote speech by Blaine Bettinger at the 2017 PMC Conference given by the Association of Professional Genealogists. She also blogged about it in her post DNA doesn’t lie! on her blog The Legal Genealogist.

Bettinger says: DNA alone can never be enough to prove a genealogical relationship.

The two cases talked about in her post are extremes, to be sure, but the principle is sound for every case. A complete stranger is just that -- a stranger.

• You don't know if they have a higher total shared cM with you because they're related to you via both the paternal and maternal side of the family.

• You don't have any generational clues because of your relative ages (and even that can be unreliable -- it's possible to have aunts and uncles who are your contemporaries if you have sibling sets that are spread out enough in time).

• Does your match come from a population where there is endogamy?

If the match is a total stranger, you don't have the information you need to analyze the results properly. The tools mentioned in the previous answers can be a guide, but the results you get from those tools aren't proof, and shouldn't be treated as such.

• Endogamy is not a factor with a 2183 cM match. This is absolutely a close family member by blood, we just don't know the exact relationship. Endogamy would not affect the cM ranges enough to change the potential relationship list. Close relative marriage (like uncle/niece or first cousins or double first cousins) would have this effect, and so would incest, but neither of these is an example of endogamy.
– Cyn
Commented Jan 3, 2019 at 17:23