# A question concerns prime numbers

I have arrived to this conjecture in my work, I am not sure that is true or false. So I would appreciate if someone give a counterexample or prove it.

My question: Let n be a non-prime such that n-1 be a prime. Is it true that there exist a prime p such that n-p is a prime as well?

-
This is exactly the special case of Goldbach's Conjecture for even numbers that are 1 more than an odd prime. – user22479 Aug 13 '12 at 7:27
Could you give some more details about the work where this question arose? – Yemon Choi Aug 13 '12 at 7:58
In Goldbach's day, the number 1 counted as a prime, hence the clean conjecture that "every even number is the sum of two (old-fashioned) primes." The modern version, namely that every even number is the sum of two (modern) primes, is stronger exactly by the gap which is the OP's conjecture, hence some natural interest. – David Feldman Aug 13 '12 at 9:33
retagged as open problem. – Benjamin Steinberg Aug 13 '12 at 14:03
Deal all thank you so much for you answers. Actually I have arrived to this question in group theory and I have never thought that is an old open problem. I would appreciate, if you let me know any approach concerns this special case. – Tina Aug 15 '12 at 10:22