Is there really a 'warrior gene' that makes people violent?

Short answer: no, not in any useful sense. The "warrior gene" is a media nickname that overstates what the actual science says.

The gene in question is MAO-A (monoamine oxidase A), which codes for an enzyme that breaks down certain neurotransmitters including serotonin and dopamine. A specific low-activity variant - sometimes called MAOA-L - has been associated, in some studies, with slightly higher rates of impulsive aggression, but only in combination with severe childhood maltreatment.

The original 2002 study by Avshalom Caspi and colleagues was important: it suggested that the variant alone didn't predict aggression, but that children with the variant who were also abused showed elevated antisocial behaviour. This was one of the first concrete demonstrations of a gene-by-environment interaction in psychiatry.

What it absolutely does not mean: that carriers of MAOA-L are warriors, violent, or predisposed to crime. About a third of men of European descent carry the variant. The vast majority live unremarkable lives. The pop-science nickname (and a famous 2009 court case where a defendant tried to use it for legal defence) ran far ahead of what the science supports.

The honest summary: a small genetic effect, only visible in extreme environments, dramatically over-marketed in headlines.

Quick reality check: this is true of basically every "X gene" media story.

The "infidelity gene," the "happiness gene," the "language gene" (FOXP2 - a real gene with a real role in speech, but again the headlines outran the data). The pattern is consistent. Find a small statistical association in one study, give it a memorable nickname, and the science gets crushed by the metaphor.

When you see "scientists discover gene for [behaviour]," your default should be skepticism. Behaviour is almost never traceable to a single gene. The interesting science is in the complicated, environment-dependent interactions - which doesn't fit in a headline.