Ancient tribal warfare resembled what we know of late North American natives' warfare: Lots of raiding (more for stealing than killing), every man a warrior if need be, reliable peace being the exception to the rule.
Gaul was largely like this when Caesar conquered it, albeit with a system of developed hierarchical loyalties resembling feudalism. The threat of raids and large scale warfare was constant, though. As a consequence, almost every boy learned to fight during his upbringing and settlements were not built with purely economical concerns in mind, but also with defence in mind: Towns and villages on hills with fortifications were normal (with moats substituting for a hill's protection in flat lands).
|Iron age people didn't choose and set up such real estate|
for settlements because of the picturesque scenery.
A generation after Gaul's conquest the final uprisings were defeated (a common chronology for Roman expansion save for Northwest Germany). The Pax Romana allowed for settlements to ditch their defence considerations, and paying tariffs and taxes replaced the general defence preparations. The economy developed favourably, towns grew without the area constraints imposed by fortifications.
Later on, the Germanic chieftains understood that the Roman Empire had a hard shell, but a soft interior: Raiding had not only become relatively less risky, but it also paid off more. Sooner or later internal struggles were bound to tip over this unstable situation. The raids forced defence preparations on the people of Gaul again; settlements in valleys were given up, fortified settlements on hills were recreated.* The economy plunged, thus the taxes and tariff revenues for the professional troops plunged and the Romans had combined the worst of two worlds.** This situation wasn't exactly sustainable either.
Luckily, we don't face any all-warrior culture today, so we're probably not going to be punished for omitting compulsory basic military training for everyone or for neglecting civil defence efforts. Today's only populations with a high degree of warrior culture are probably the Tutsi and Israelis. The Chechens gave it a try and found it to be a faulty approach.
Still, the trade-off between defence preparations and economic optimisation is interesting (to me). The example above goes well beyond a mere military spending issue. It's about whether or not an entire society is coined by discouraging aggression and warmaking capability.
Amazingly, the entire Western World navigated through the existential perma-crisis known as the Cold War without turning into all-militarised societies. The Soviets gave it a try, with allegedly up to a quarter of the economic activity of the Soviet Union being for the military strength (its economy was heavily subsidised by unfair trade with its satellite countries, though). They lost. This time the 'hard shell, soft interior' approach prevailed over the perma-war approach.
Maybe this economy-emphasising strategy is superior. The Romans defeated the Gauls, after all. No all-warrior culture lasted as long as the Roman hard shell/soft interior culture did (The East Roman empire shrank, but it survived till about 1,600 years after Rome de facto gave up the all-warrior approach.)
- - - - -
Now to another aspect (this should probably be two blog posts):
I mentioned the important differences between distributing new income and redistributing already distributed income before.
The economic growth which occurs when a society becomes unshackled from the distortions of defence preparations was easily distributed. Much of this was distributed away from defence; it did not feed the professional military establishment. A redistribution towards military efforts was hard, much harder than the initial distribution of the growth. The Romans experienced additional economic troubles because long-distance trade collapsed due to political instability and the insecurity of trade due to civil wars. It's most hard to redistribute/reallocate resources if the resource base is even shrinking. All interested parties will fight tooth and nails to keep their share of the pie, even as it's shrinking anyway.
Maybe it would have been wiser to maintain huge efforts in large construction projects, such as new roads, new Colosseums, pyramids et cetera. The resources allocated to such chronic public investment resources could have been transferred to military efforts in times of need and easily so.
The modern method to overcome the resistance to reallocation of resources / redistribution of income is different: We hand out bonds instead of increasing tax revenue extremely. The difference is probably a mere wartime illusion coupled with a very real post-war redistribution of income towards the pre-war wealthy, but it appears to be effective.
*: Some such reversions occurred only later, when tribal warfare returned in full force instead of earlier when raiding began anew.
**: A late medieval Chinese policy in face of Japanese raiders ("pirates") was even worse; the coastal regions were given up, settlements there were outlawed in order to deprive the raiders of targets.
P.S.: The errorists' success in provoking lots of defence precautions such as airport security etc. may be an analogy to raiding Germanic warbands in the late Roman empire, but I still don't think that they're going to reach a critical mass..