Writing is usually only the end process of a series of other processes, when the theme under consideration begins to take shape. Even writing a chatty letter requires some forethought and planning, albeit perhaps quite unconscious; the assembling of thoughts and opinions, items of news, description of feelings, all poured out in whatever order the writer feels appropriate. For a book, the planning stages can take more time than the actual writing, and when writing begins, there are constricting factors like word limits, deadlines, and the need to impose some sort of order on a multiplicity of unrelated facts. The other constricting factor is that the day jobs, which pay the mortgage and buy the food, tend to get in the way. All writers should have a housekeeper, cook and cleaner, chauffeur, garage mechanic, computer repairer, and native bearers to take the books back to the library before they accumulate crippling fines. And then there is the factor of advancing age, that leaves only the energy to perform aforesaid day job and then slump in front of Star Trek videos until it is time to go thankfully to bed, only to be forced into worried wakefulness at the proverbial three in the morning, because there are only six months left before the deadline.

At the outset of the necessary research to provide the data base, the writer must find a way into the subject, and quickly identify the primary sources. For classics, ancient history and archaeology, much of the hard work has been done by the 19th century compilers and scholars. There are, nowadays, many translations of ancient works, and the list is growing all the time. There are copious, learned assemblages of inscriptions, sites and monuments records, and archaeological excavations. Not all of these will apply to the subject being studied, so the necessary extracts must be made. This is not always as easy as it sounds. For one thing, the source material will not necessarily be conveniently located in one library, so it is a good idea to inveigle your way into membership of as many libraries as possible. Once installed in a library, the problems do not resolve themselves easily; some of the items are too large to photocopy, or they are not on open access, so you have to ask for them to be brought to you, sometimes after a wait of a few days, not because the librarians are obstructive, but because the items come from a completely different site. These books are usually not available for loan, so you cannot take them home to read at your leisure. The one commodity that any writer must be prepared to expend liberally is time. This is where the day jobs really get in the way. Unless the writer is fortunate enough to have a sabbatical or other free time, he or she must learn to make best use of time. Good organisation is vital. Arriving at the library without the list of references carefully prepared the night before could be classified as a complete waste of time, unless you turn it to advantage by scanning the contents of lots of periodicals for anything you may have missed. The depressing thing is that on occasion a seminal article can be discovered in this way, not mentioned in the bibliographies of the books you have read so far.

Having found a way into the subject, the writer usually faces one of two problems: there will be a dearth of information, or there will be far too much. The former case involves much re-reading and floor pacing, with the problems constantly at the forefront of the mind; writing down sudden questions or general points on the backs of envelopes or supermarket till receipts; thinking of things in the car at 70mph on the motorway and wishing for a recording device on the dashboard; missed meals or indigestion after meals consumed inattentively. The lack of primary sources must be compensated by deep thought and the sifting of the opinions of other people, who have written the secondary source material; quite soon one's own opinions begin to develop, surprising sometimes in their table-thumping, adrenaline-surging intensity. Without opinions no-one can write anything, even if those opinions are not starkly outlined in the resultant text. Sometimes the idea is so evolved and all-pervading by the time it reaches the page that the writer assumes that all readers must know absolutely everything that he or she takes for granted, and therefore can't make it clear because it is not even realised that there is a need to do so. Most readers have probably experienced at least once the agonising 'Yes, but what are you saying?' after struggling through masses of words.

In the cases where there is a great deal of information, both primary and secondary, the task becomes one of patient labour, classifying into sub-sections, cross-referencing, and then evaluating the material. The maxim 'Don't believe everything you read' should be tattooed on the back of whichever hand the writer uses to lift the coffee cup. It is easy to be swayed by clever arguments and then find yourself contradicting in chapter four what you said in chapter two.

The huge volume of material on Augustus, in several languages, ancient and modern, has been described as 'daunting' which comes nowhere near to conveying the terror of confronting it. The Thesaurus on my computer offers 'frighten, shake, cow, appall, dismay, intimidate, overawe, subdue, frustrate, dishearten, discourage, dispirit, thwart'. Yes; all those things apply. With so many books and articles already extant, why bother to write anything more? That would have been a reasonable excuse to slip out of the contract and pay back the advance, vanishing into the sunset with an overriding sense of relief and several Star Trek videos clutched under each armpit.

But if you've started you may as well finish. The mass of material on Augustus usually takes up only one theme, however broad that theme may be. The art and architecture of the period, the coinage, the various political tangles, the civil wars, the foreign wars, the development of the army, the growth of the administration, the adaptation and application of the law, have all been placed under the academic microscope and been pronounced upon and argued about. The books that deal with the epoch quite often follow a chronological course until the so-called first settlement of 27BC and then take up the thematic approach, with chapters on all the above mentioned topics and more narrowly specific ideas. Somewhere among all this there is the life of a man who entered history as a teenager and remained in command of the Roman world until he died long after the modern retirement age. The thematic approach obscures the time factor, and also the ad hoc adaptability to circumstances that was Augustus' strongpoint. Even when dates are provided in a discussion of a theme or a related series of events, the close juxtaposition of facts all logically assembled in the space of a few pages makes it seem as though there was a master plan for the Empire which Augustus had already outlined for himself right from the beginning and put into operation stage by stage. Several authors have stated that this was not the case, but nevertheless, by reading brilliantly researched accounts of the various themes underpinning the developing Roman Empire, the impression of a divinely pre-ordained greatness for Augustus is merely strengthened.

In the face of so much research on Augustus it is impossible to find an untrodden path. The difference in this latest book is one of approach and organisation of material. The book is strictly chronological, attempting to meet events as Augustus himself met them, as people do in real life, without the benefit of hindsight that inevitably casts a different fight on what went before. The arrangement of material in this way necessarily involves some repetition of facts, as Augustus dealt with various matters successively; some scene setting to explain what he was doing is necessary each time the subject comes up for consideration. The approach has the advantage of highlighting the length of time it took to establish the final form in administering such a large extent of territory with such a diverse population. Some elements never reached a final form in the reign of Augustus, but merely evolved slowly, undergoing changes here and there, still requiring adaptation in the reigns of successive Emperors. This continual evolution was a feature of Augustus' rule of the Roman world; he was not a man to insist upon an established method if circumstances dictated that changes should be made, but at the same time in making changes he did not usually force through the necessary measures, flattening all opposition. When changes came, they were often the result of experimentation in persuasion and example, so that the Senate and People of Rome either condoned the ideas or thought of them as their own.