According to Martin Goodman's (and Jane Sherwood) 'The Roman World 44BC-AD180, Augustus kept such a large army mainly to secure power. He mentions how previous armies had been much smaller during Republican times and also how many of the legions saw very,very little action,e.g. legions in Egypt had almost nothing to do from 24BC to AD170. He also points out that Roman strategic defense was more about strong retaliation rather than quickly reaction or sealing off the the area of attack. Any users who could give an indication of how true is this argument is?
This isn't a particularly compelling explanation of what data we have, I would argue for two main reasons: firstly, one of Augustus' first acts was to slash the size of the Roman military in half. The imperial army appears larger than earlier Republican ones primarily because it was controlling significantly more territory. In fact, in a per capita sense the Roman army was only probably only 0.5% of the entire population, which is a drastic reduction in the population under arms during the pre-Roman period (this led to what is called a "guns and butter dividend", as essentially the amount of money that would be spent on the army could go to construction, trade goods, urban economies, etc). So, the first issue is that there is no unusually large army requiring an explanation, as the empire was actually fairly shoestring.
Similarly, the idea of the army being primarily for internal suppression simply does not match the data as to where the army was deployed. Overwhelmingly, almost exclusively, the legions were stationed on the frontiers, leaving the interior of the empire essentially demilitarized. Seeming exceptions to this actually prove the rule: the legion(s) in Spain guarded the restive mountains of the Vascones (Basque) and the gold mines, the legion in Wales also was dealing with an "internal frontier" of the welsh highlands. And if you look more closely at, say the placement of camps in Dacia, you see that they ring the mountainous regions and the northern frontier and leave the vast lowland interior practically empty. So to put it simply, if the legions were intended to suppress internal dissent they were deployed in a way that was remarkably bad for that.
So what did they do? Well, we can actually look at what the legions do in Egypt because of documentary records (incidentally, there were two legions, so 10-20,000 men, in a province of 5-7 million). certainly, the main base was near Alexandria, a historically troublesome city. A lot of what they did was guarding against desert nomads and bandits--in fact, to a much greater extent than in other provinces because the Romans adopted the police system that went back to pharonic times. Internal suppression was instead left to local communities.
I don't agree with this argument at all, and I even find it a tad strange. The argument about the army size may be true if you don't look at any other facts, but you can obviously not do that. While the army sizes indeed where smaller during the pre-Marian parts of the republic, they certainly were not small by the end of the republic (during the civil wars). At the end of the civil wars, the legions amounted to ~60 in total which had a huge strain on the economy, not to mention that it was rather useless to keep an army that big when it wasn't needed. Augustus recognized this, and shortly after Actium he reduced the total amount of legions in the empire to half that number (Res Gestae, 3; Tac. Ann. 4.5).
About the amount of action that the different legions saw, it can be said that in hindsight that could have been improved, but at the time it was necessary to keep legions around in case rebellions would break out. Constantine seemingly realised this when he came to power and made adjustments, which also could be judged in hindsight. Keep in mind that Egypt was a very rebellious region, which overthrew its previous ruler Cleopatra in favour of her brother (Caes. Civ. 3.103 onwards). That, combined with the cheer size of Egypt, and more importantly, Rome's dependancy on Egypt's food production, made it reasonable to keep three (mostly idle) legions there.
I'd be interested in seeing the source they use to determine that Rome's strategy was to retaliate rather than defend, and the argument for it. While situations definitiely can be found where that is true (Carrhae seems a good example), they don't show upon a strategic choice to rely upon retaliation, but rather a lack of ability to quickly communicate over distances and thence also reinforce. While we have sources that show us the amazing speed at which Roman legions could march from Gaulish to Italic provinces, we're still talking about days for such distances with fully mobilized legions. If you look at Egypt, or at the legions lost by Varus in 9AD, you can clearly see that they used strategic defense. We also know that legions where moved around a lot (relatively) in the empire, which would suggest that they were more interested in proactively protecting the provinces (and controlling them) than having to retake them from the enemy.
With that said, let's move on to why I think the argument that Augustus (or the Julio-Claudians even) kept such a large army mainly to secure power, is wrong. What I've written above should show you that the army size in the empire hardly was excessive, but even so the Julio-Claudians actually controlled the army, which can't be dismissed completely. It should be said that, given the necessity, the army could have been used to maintain Julio-Claudian power in the empire. However, with the lack of any such situations during Julio-Claudian rule, the whole argument that it was used for that, is moot - in my opinion. We know of rebellions that were fought down during the early imperial period, especially more distant provinces tried to break free of Roman rule. It would be a stretch, to say the least, to use the fact that the army was used to beat back outright rebellions in order to form an argument for that the army was used mainly to secure power.
The Julio-Claudian power did not come from their control of the army, but was a product of both past events and intelligent propaganda from the imperial family. Following the assassination of Julius Caesar, the Second Triumvirate (Octavian, Mark Antony and Lepidus) compiled a list of proscriptions against the people guilty of assassinating Caesar. However, that list would only have been some 20 names long, but it also included names of more or less anyone who supported the 'wrong' political camp, which meant that the list included a lot of other people as well, well into the hundreds.
Anyway, the removal of most of these people who were opposed to Caesar, along with the general popularity he had with the people and gifts given to the people from his will, meant that the Second Triumvirate had a solid foundation during the civil war against Brutus and Cassius. After the next civil war - between Octavian and Mark Antony - Octavian had managed to gain even more popularity with the people and was thus in rather good standings 'at home'.
Jumping past events like Cleopatra's suicide and Octavian's proclamation as Augustus, some interesting things can be seen in the behaviour of primarily Augustus and Livia. As it happens though, I wrote a larger post on a similar topic a couple of months ago, so rather than repeating myself I'll leave links to them here:
Part 1
Part 2