We want to start with asking what the distinction is between soldiers and warriors. It is a tricky question and even the U.S. Army sometimes gets it badly wrong ([author Steven] Pressfield, I should note, draws a distinction which isn’t entirely wrong but is so wrapped up with his dodgy effort to use discredited psychology that I think it is best to start from scratch). We have a sense that while both of these words mean “combatant”, that they are not quite equivalent.

[…]

But why? The etymologies of the words can actually help push us a bit in the right direction. Warrior has a fairly obvious etymology, being related to war (itself a derivative of French guerre); as guerre becomes war, so Old French guerreieor became Middle English werreior and because that is obnoxious to say, modern English “warrior” (which is why it is warrior and not “warrer” as we might expect if it was regularly constructed). By contrast, soldier comes – it has a tortured journey which I am simplifying – from the sold/sould French root meaning “pay” which in turn comes from Latin solidus, a standard Late Roman coin. So there is clearly something about pay, or the lack of pay involved in this distinction, but clearly it isn’t just pay or the word mercenary would suit just as well.

So here is the difference: a warrior is an individual who wars, because it is their foundational vocation, an irremovable part of their identity and social position, pursued for those private ends (status, wealth, place in society). So the core of what it is to be a warrior is that it is an element of personal identity and also fundamentally individualistic (in motivation, to be clear, not in fighting style – many warriors fought with collective tactics, although I think it fair to say that operation in units is much more central to soldiering than the role of a warrior, who may well fight alone). A warrior remains a warrior when the war ends. A warrior remains a warrior whether fighting alone or for themselves.

By contrast, a soldier is an individual who soldiers (notably a different verb, which includes a sense of drudgery in war-related jobs that aren’t warring per se) as a job which they may one day leave behind, under the authority of and pursued for a larger community which directs their actions, typically through a system of regular discipline. So the core of what it is to be a soldier is that it is a not-necessarily-permanent employment and fundamentally about being both in and in service to a group. A soldier, when the war or their term of service ends, becomes a civilian (something a warrior generally does not do!). A soldier without a community stops being a soldier and starts being a mercenary.

Incidentally, this distinction is not unique to English. Speaking of the two languages I have the most experience in, both Greek and Latin have this distinction. Greek has machetes (μαχητής, lit: “battler”, a mache being a battle) and polemistes (πολεμιστής, lit: “warrior”, a polemos being a war); both are more common in poetry than prose, often used to describe mythical heroes. Interestingly the word for an individual that fights out of battle order (when there is a battle order) is a promachos (πρόμαχος, lit: “fore-fighter”), a frequent word in Homer. But the standard Greek soldier wasn’t generally called any of these things, he was either a hoplite (ὁπλίτης, “full-equipped man”, named after his equipment) or more generally a stratiotes (στρατιώτης, lit: “army-man” but properly “soldier”). That general word, stratiotes is striking, but its root is stratos (στρατός, “army”); a stratiotes, a soldier, for the ancient Greeks was defined by his membership in that larger unit, the army. One could be a machetes or a polemistes alone, but only a stratiotes in an army (stratos), commanded, presumably, by a general (strategos) in service to a community.

Latin has the same division, with similar shades of meaning. Latin has bellator (“warrior”) from bellum (“war”), but Roman soldiers are not generally bellatores (except in a poetic sense and even then only rarely), even when they are actively waging war. Instead, the soldiers of Rome are milites (sing. miles). The word is related to the Latin mille (“thousand”) from the root “mil-” which indicates a collection or combination of things. Milites are thus – like stratiotes, men put together, defined by their collective action for the community (strikingly, groups acting for individual aims in Latin are not milites but latrones, bandits – a word Roman authors also use very freely for enemy irregular fighters, much like the pejorative use of “terrorist” and “insurgent” today) Likewise, the word for groups of armed private citizens unauthorized by the state is not “militia”, but “gang”. The repeated misuse by journalists of “militia” which ought only refer to citizens-in-arms under recognized authority, drives me to madness).

(I actually think these Greek and Latin words are important for understanding the modern use of “warrior” and “soldier” even though they don’t give us either. Post-industrial militaries – of the sort most countries have – are patterned on the modern European military model, which in turn has its foundations in the Early Modern period which in turn (again) was heavily influenced by how thinkers of that period understood Greek and Roman antiquity (which was a core part of their education; this is not to say they were always good at understanding classical antiquity, mind). Consequently, the Greek and Roman understanding of the distinction probably has significant influence on our understanding, though I also suspect that we’d find distinctions in many languages along much the same lines.)

Bret Devereaux, “Collections: The Universal Warrior, Part I: Soldiers, Warriors, and …”, A Collection of Unmitigated Pedantry, 2021-01-29.