Lands of Red and Gold

Status
Not open for further replies.
You know, your depiction of the Waiting Death is a very dark one, but you may be missing out on the other side of the coin. Quite a few societies have developed celebrations and rituals that make light of death. Ancient (and modern) Mexico, for example. I could see a culture developing in which it is customary for a person infected to act very differently during their period of "waiting."

Very good point! There is of course some uncertainty involved in the process, but if someone knows that there's a good chance that they could be dead in two months, they might start making arrangements. All sorts of arrangements.

Some cultures might expect victims to spend the period in a state of religious ecstasy, and assume that the ones who died did so out of a failure in piety. Others might seek to model virtuous and unassuming lives. You might find some where the infected run to and fro, proclaiming love and reconciling grievances, with the culture at large viewing the period as a transitional one in an individuals life - a liminal state where standard societal taboos are relaxed. Hrm. Pseudo-berserkergang aborigine warriors? A month can be a fairly long period in wartime....

Yes, all of those would be possibilities, depending on the culture and the individual. If none of your friends knew that you caught Marnitja, now might be the time to start borrowing large sums of money... In general, I'd expect that there was a tradition of settling your affairs in case of death, although there'd be all sorts of other options. The idea of berserkergang is particularly appealling. This could link into their idea of animal totems, and of course they have a couple of drugs which might help to achieve the proper state of mind.

Heaven help the nation whose king thinks he has to establish his legacy right now.

"May you live in interesting times..." In those cultures where there's no automatic law of primogeniture (which is, in fact, pretty much every culture in *Australia), the squabbling which broke out once the king had contracted Marnitja would be murderous in every sense of the word. And if the king decides to go out in a blaze of glory, things could get very bloody indeed.

As for the emus, people seem to be missing the obvious somehow. There's a large, monied minority waiting in Europe for a new source of meat, and a reason Israel is number one in the world for per capita consumption of Turkey.

I thought about that, but sadly I suspect that emus will be ruled non-kosher. With birds, apparently the rulings are made on a case-by-case basis, and the guiding principle is similarity to other birds. Turkeys were deemed to be kosher because they were similar to a bird in India which was thought to be kosher. Emus would probably be decided to be closest to the ostrich, which is a non-kosher bird, apparently (although I've seen a couple of sources which differ on that). Although there's always the possibility that some rabbinical authorities may rule otherwise, I suppose.

IIRC a critter had to be associated in some fashion with water ( so people could accept it as a sort of fish) to be eaten on Lent . Muskrats in NA and capybaras in SA fit the bill.Although I doubt the Vatican ever officially ruled on the subject.

So maybe Australian ducks, but emus would be too much of a stretch.

On the subject of Maori/Aborigine contact, do you see any religious ideas or cultural practices traveling between the two cultures?

Hard to say. The contact will be long-range and intermittent, but there will be a few people who migrate permanently from one land to the other, and they may bring some ideas with them. It will get particularly complex because the Maori religious beliefs are still forming in New Zealand; they've only been inhabiting those islands for a few decades when they first make contact with Australia. So it's possible that some religious ideas will seep through.

What makes it even more hard to judge is that the major organised religions in *Australia - those with established priesthoods and written texts - are a long way away from where the Maori first make contact. Their landings are around the Illawarra, Sydney, and the southern coast of New South Wales. Those places are relative backwaters, in cultural terms - a long way from sources of bronze, and with only a relatively low population.
 
By the time Europeans arrive on the scene can we expect an Australia of City-States with Homeric Warrior-Kings fighting to carve out their own personal empires? :)
 
Can't believe I haven't commented on this yet. Great job Jared. Perhaps we can forgive you for not making this as world-covering as DoD after all...

In regards to Uluru (sp?), is it going to become a religious part of Aborginial life? The idea of some faction carving a temple of sorts into the side of the giant rock would just be epic.

Finally, do you know exactly how far this TL will cover yet? Or is it a wait-and-see kinda deal? Thanks and keep up the good work.
 
By the time Europeans arrive on the scene can we expect an Australia of City-States with Homeric Warrior-Kings fighting to carve out their own personal empires? :)

There is a land, the land of bronze, the land of mist, the land of courage, where courageous Tjunini soldiers battle endlessly with crafty Kurnawal warriors, where the wild men still lurk in the highlands, raiding where they may, and where in the long winter evenings men gather to feast around roaring fireplaces, drink endless goblets of gum cider, and hear the bards recite the endless verses of the War of the Princess, and even the smallest boy can recite the names of every captain who led men into that war, while in the courts of the kings poets compete with each other to create ever more complex verses packed with allusions and circumlocutions which only the most learned of listeners can fully grasp...

Or, to put it another way, depends very much on which part of Australia you're talking about. On the one hand, there are parts of Australia which still don't even grow crops. On the other hand, there are parts which are in the Post-Imperial era at the time of European contact. And much in between.

Can't believe I haven't commented on this yet. Great job Jared. Perhaps we can forgive you for not making this as world-covering as DoD after all...

Heh. Think of it this way. DoD didn't require me to write the history of the world to start with. This one does, or at least the history of a continent. It's been getting a bit... wordy.

In regards to Uluru (sp?), is it going to become a religious part of Aborginial life? The idea of some faction carving a temple of sorts into the side of the giant rock would just be epic.

Uluru is a long way out into the desert, as far as the settled peoples are concerned. I'm not sure whether the locals will be able to do something, but for those who are long way away... well, it would be a long pilgrimage, to say the least. Religion gets expressed in a variety of ways, from temples to pilgrimages to feasts to, well, you really don't want to know what goes on inside an Atjuntja temple.

Finally, do you know exactly how far this TL will cover yet? Or is it a wait-and-see kinda deal? Thanks and keep up the good work.

This TL is pretty much a "write it and see what happens" deal. Unlike DoD, where I planned things a very long way in advance, I know only a few very broad trends about the world and particular countries. As I've mentioned before, the TL focuses almost exclusively on Australasia. The one possible exception is that I may subcontract out some writing of other parts of the world, for the first couple of generations post-1619, at least. The effects of Australian contact are going to cause all sorts of butterflies in the Thirty Years War, fer'instance, and I lack enough knowledge of that period to work out exactly how things are going to play out. But I haven't decided anything definite about that yet.
 
The one possible exception is that I may subcontract out some writing of other parts of the world, for the first couple of generations post-1619, at least. The effects of Australian contact are going to cause all sorts of butterflies in the Thirty Years War, fer'instance, and I lack enough knowledge of that period to work out exactly how things are going to play out. But I haven't decided anything definite about that yet.

Jared, can I just say for one that if you cause an outbreak of Marnitja in Europe during the Thirty Years' War, I will personally worship you as a god. Because any TL where you can kill that damned Swede Gustav Adolph with a disease that eats your lungs...priceless.

And I guess cannibalistic orgies for the Atjuntja temples, out of sheer weirdness? Do I get a cookie?
 
Lands of Red and Gold #9: The First Speakers
Lands of Red and Gold #9: The First Speakers

The time of the Collapse was one of great panic and greater upheaval. Harvests failed, droughts persisted for years and stretched into decades, bushfires grew more frequent and spanned ever greater areas of the continent, and the arts of civilization seemed to failing. In this time of chaos, people sought refuge in whatever consolation they could find.

In 842 BC, Robinvale, one of the old Wisdom Cities, was a place ripe for new ideas. Uprising and subsequent collapse had destroyed the great city of Murray Bridge, further downriver. In turn, this had ruined the arsenical bronze trade which supplied most of Robinvale’s wealth. The seemingly endless drought had destroyed much of its agricultural hinterland. More than half of its former territory had already been abandoned, wattles and yams left to run wild while wetlands silted up and returned to semi-desert. Nomadic hunter-gatherers reoccupied the former farmlands, while the surviving farmers gathered in Robinvale itself, growing increasingly unruly.

Wunirugal son of Butjinong was one such farmer among thousands who arrived in Robinvale in that year. The exact location of his old farm is not known; at least two dozen sites would later be claimed to be the site of his birthplace. He is reliably known to have been a member of the Azure kitjigal, and he claimed the wedge-tailed eagle as his personal totem. Beyond that, nothing definite is known about his life before he arrived in Robinvale in the summer of 842 BC, although a thousand tales have since sprung up to explain how he spent his early years.

Just as with his early life, what Wunirugal accomplished in Robinvale in that year has been recorded in many contradictory versions. Certainly, Wunirugal was among the more vocal of the farmers complaining about the lack of food, to the point where the city militia took official notice of him. Some say that he spoke so eloquently that the militia agreed to escort him before the Council to plead his case. Some say that Wunirugal outran the militia and entered the Council hall on his own. Some say that he was struck immobile by a vision and was carried bodily into the Council hall for investigation. One account claims that the Council came out to meet him in the main square of Robinvale, but that version is usually discounted.

No two versions of Wunirugal’s meeting with the Robinvale Council agree on what he said, or even on the names of the members of the Council. There is surprising unanimity about the Council’s decision: Wunirugal was deemed a danger to public order, ordered to be expelled from Robinvale immediately, and not to return for a year and a day, on pain of death. Wunirugal left, but he was not silent along the way. Again, the accounts of his words vary, but all sources agree that he persuaded at least three hundred people to come with him into exile.

Wunirugal led his new followers far from Robinvale. Accounts of their journey include some fantastic events. The most nearly-universal of those events is an account of how soon after he left the city, Wunirugal received another vision. While having this vision, he was struck by lightning which came from clouds which produced no rain, yet he survived with no ill effects. Scholars will long argue whether this widely-reported event is factual. It might be; some lightning strikes from thunderclouds where rain is falling but evaporates before it reaches the ground, and some people do survive lightning strikes. Certainly, the reports of what Wunirugal did after the lightning strike are in surprising agreement. He is said to have fallen to his knees and said: “Help me, help me. Tell me what I should do, O lightning blue?”

Whatever the merits of this account, it is clear from the many tales of Wunirugal that he had visions, or what other people believed to be visions. He drew on these for guidance, and led his followers up the Murrumbidgee. This river is one of the major tributaries of the Murray, but it had been a relative backwater since most of the trade flowed up and down the main river. Wunirugal led his followers past a long series of natural wetlands, and reached an area of lush, fertile soils which flourished even in the drought. He declared that this would be the perfect place to build a town. According to most (but not all) accounts, he said, “Here, the earth will always grow. Here, we can build a city which will have no rival.”

He called the new city Garrkimang.

* * *

Garrkimang [Narrandera, New South Wales] grew to become one of the four great cities of the Classical Gunnagal. Like their contemporaries, the people of Garrkimang could trace their ancestry back to their Formative forefathers. However, Garrkimang developed along a different path from its neighbours. In language, it was much more distinct than any of the other Classical cities. The migrants who founded Garrkimang were a combination of refugees from the former Murray Bridge and the most westerly areas of former Robinvale territory. This meant that their speech had diverged much further; while there were clear underlying similarities with other Gunnagalic languages, learning the speech of Garrkimang was considerably more difficult for foreigners than that of any other Classical city.

In culture and religion, Garrkimang was also unlike any other Classical city. The inhabitants called themselves the Biral, a name which means roughly “chosen ones.” They traced this back to the migration under Wunirugal, believing that they had been chosen to be granted their new land as a sacred trust. Their religion had a similar foundation to the older Gunnagal beliefs; they still shared the same general view of the Evertime and of the spirit-beings who inhabit eternity, although they gave different names and attributes to many of those beings. Yet the old beliefs had been overlaid by a new religious structure, that of the First Speakers and their representatives who interpreted the world.

The heirs of Wunirugal ruled Garrkimang as absolute monarchs. The old cities had been ruled by oligarchic councils, but there had never been such an institution in Garrkimang. The rulers claimed the title of First Speaker, and based their rule on religious authority. They asserted that they were entitled to rule because they possessed the talent of interpreting the wisdom of eternity. They proclaimed their rule through a series of law codes first promulgated in Garrkimang itself, and which were spread throughout every city and town which came under their rule. They also adopted a set of protocols in terms of conduct, dress, and ceremonies to support the view of the First Speaker as the greatest moral authority. This was most obviously shown in the privilege which gave the First Speaker his title: in any meeting or ceremony, the First Speaker always was the first person to speak. Anyone who dared to speak to the First Speaker unless directly addressed first would be fortunate if they were simply exiled; death was a common punishment for such a social faux pas.

The First Speakers were not always direct heirs of Wunirugal; the succession was open to all males in the royal family. It was even possible, although difficult, for those not of direct royal descent to be accepted into the House of the Eagle; adoption was an accepted method for particularly eminent people to join the royal family and become eligible for the throne. The succession was often decided by the will of the First Speaker, who would designate an heir from amongst his relatives. On some occasions, the royal princes would meet to acclaim an heir when the succession was unclear. Public disputes over the succession were rare, and civil wars over succession would be unknown until the declining days of the monarchy. Incompetent rulers could, however, be removed. If the royal family thought that a First Speaker was very bad at listening to the wisdom of eternity, then that First Speaker would be quietly offered an opportunity to commune with eternity more directly, and another member of the royal family would take the throne.

Despite some internecine intrigues in the House of the Eagle, Garrkimang’s monarchs were always much more secure on their thrones than the rulers of any other Classical city. The royal family had the bastion of religious authority to support their rule. More than that, the old kitjigal system had broken down during the time of the Great Migrations. With so many people displaced, two of the eight kitjigal were lost entirely, and the rest were abandoned as social institutions. In the other Classical cities, the kitjigal evolved into armed factions which preserved their own privileges, including the right to form social militia. Monarchs in Tjibarr, Weenaratta and Gundabingee always feared uprisings amongst the factions, but Garrkimang did not have this threat.

Some aspects of the kitjigal were still preserved in Garrkimang, but in much-changed form. From its founding, Garrkimang’s armies were traditionally divided into six warrior societies, each of which had their own initiation rites, values, informal social hierarchies, and special duties. These societies were named the Kangaroos, the Corellas, the Ravens, the Kookaburras, the Echidnas, and the Possums. Each of these societies derived their names and some of their values and practices from the old colours and social codes of the kitjigal [1]. Garrkimang also had six trading associations, each of which emerged from the old kitjigal colours. These formed into a system of recognised partnership and profit-sharing, and were in effect early corporations which had collective ownership of farming land, mines, trading caravans, and other ventures, who shared the profits and risks amongst all members of their society. The trading societies were powerful voices within Garrkimang and its dominions, but since all military and religious power was reserved for the monarchy, the trading societies never acquired the same political power which the factions did elsewhere.

* * *

Garrkimang occupied what was probably the best agricultural site of any Classical city. It had the convenience of large areas of productive land upriver suitable for the Gunnagalic system of dryland agriculture, and a series of natural lagoons and other wetlands downriver which were easily expanded into the managed artificial wetlands which the Gunnagal so favoured. The wetlands downriver of Garrkimang were productive enough that the First Speakers encouraged the diversion of some water to irrigate a few chosen crops, unlike the usual Gunnagal farming system which relied on rainfall. This irrigation was mostly for their favoured drug, pituri, but also for the cultivation of a few fruits and other high-status foods [2].

With productive lands as the foundation of their power, the First Speakers turned Garrkimang into the capital of a large kingdom which controlled three significant rivers, the Murrumbidgee, the Lachlan, and the Macquarie. They controlled almost all of these rivers, except for the farthest downstream areas of the Murrumbidgee and the Lachlan, which were under the control of the kingdom of Tjibarr. By edict of the First Speaker, in 256 AD the kingdom became known as Gulibaga, the Dominion of the Three Rivers.

While Gulibaga was a powerful kingdom, for most of the Classical era it was only one nation amongst four. The three Murray kingdoms of Tjibarr, Weenaratta, and Gundabingee all flourished during this era. The four nations each had a vested interest in ensuring that none of their rivals grew too powerful, which was reflected in a fluid system of alliances that prevented one kingdom from completely defeating any of the other four.

The alliance system broke down in the Late Classical period, thanks to the social disruptions of the first blue-sleep epidemics in the mid-fourth century AD, and a more than usually faction-ridden aftermath in the kingdom of Tjibarr. This let the First Speakers extend their control to the River Darling, a major tributary of the Murray, and the key transport route for tin from the northern mines. Gulibaga kept its control over the tin trade from this time, despite efforts to dislodge its forces. While tin was still traded further downriver to the other Classical kingdoms, Gulibaga received the largest share, and from then on they had better access to bronze than their rivals.

It was an advantage they would put to good use.

* * *

In the vanished era of the Classical Gunnagal’s ancestors, war was as much a series of raids for honour as it was a contest between nations. The era of the Collapse changed that; wars were now fought for national gains. Still, while Classical military tactics were more organised than those of their ancestors, they were not particularly advanced. The archetypal Classical warrior carried a wooden shield and a bronze-tipped spear, sometimes also with a short sword. Armour was rare, save perhaps an emu leather helmet. Captains might have more bronze armour, but the common Classical soldier was only lightly-protected. Battle tactics and training were not particularly advanced; being a soldier was a part-time occupation for most people, and while the Classical Gunnagal knew how to form a line of battle, their coordination and discipline were both limited.

With a near-monopoly on bronze, Gulibaga’s warriors changed the old pattern. The kingdom had the wealth and the resources to equip their leading warriors with better armour, typically a bronze helmet and greaves, and hardened leather breastplates. They could afford to maintain the first large professional standing army, elite units who trained and deployed together. They standardised and extended their tactics with a number of military innovations. Professional Gulibagan warriors carried pikes and rounded shields, transforming them into Australia’s first heavy infantry, who could break almost any enemy line of battle. In battle, the core infantry were supported by lightly-armed skirmishers who used bows with stone or bone-tipped arrows, and who helped to disrupt enemy formations.

By the mid fifth century AD, Gulibaga’s military organisation was clearly superior to anything developed by its Classical rivals. The combination of better arms, armour and tactics would prove almost irresistible.

* * *

With its superior military organisation and resources, Gulibaga transformed itself from a kingdom into an empire. In the name of the First Speakers, its armies waged war on its classical rivals, particularly its most powerful opponent, the kingdom of Tjibarr. In a series of campaigns from 467 AD to 482 AD, Gulibagan armies conquered most of Tjibarr’s territory, although the city walls withstood siege after siege. In 486-488 AD, a long siege finally broke through the walls of Australia’s most ancient city.

In previous wars between Classical kingdoms, similar victories had seen defeated monarchs being reduced to effective vassals, with “advisors” from the victorious kingdoms dictating policy. Such advisors were usually thrown out within a few years, with the support of one of the other four kingdoms. With the defeat of Tjibarr in 488, however, the First Speakers did something unprecedented: they deposed the old monarchs and created a new province with an appointed governor. This action is usually taken to be the start of the Imperial era in the history of the Murray basin. Some authorities use a later date of 556 AD, when Gulibagan armies subdued the forces of Gundabingee, the last surviving Gunnagal kingdom. After this victory, the First Speaker renamed his nation to Watjubaga, the Dominion of the Five Rivers [3]. This would be the name by which it would be remembered.

With the resources of the Five Rivers at its command, Watjubaga expanded into Australia’s first and largest indigenous empire. Its core territory remained the old Gunnagal lands along the Murray and Murrumbidgee, but its armies carried its rule to most of the agricultural regions of south-eastern Australia. In the north, one of its major early accomplishments was the gradual expansion along the Darling until they conquered the New England highlands directly, taking over the sources of tin and gems. To the south, they faced some determined resistance from the Junditmara peoples, who had their own developing kingdoms and a hierarchical social code based on duty to one’s elders, conformity, and rewarding loyalty. Still, the might of professional discipline and imperial bronze saw the Junditmara kingdoms defeated one by one.

* * *

At its height around 850 AD, Watjubaga claimed suzerainty over territory which stretched from the Darling Downs in the north to Bass Strait in the south, and to the deserts and the Spencer Gulf in the west. These northern and western borders represented what amounted to its natural frontiers. In the north, the Darling Downs were inhabited by a set of feuding Gunnagalic peoples who dwelt in small villages and raided each other for emus and honour. Their northern limits were bounded by the growth of the red yam, which does not grow properly in the tropics. The Empire imposed its authority on these peoples, although the distance and the fractious nature of its subjects meant that its authority was perforce rather loose.

Likewise, the western and southern borders of the Empire were largely bound by desert and the seas. Watjubaga controlled all of the thinly-inhabited lands west of the Darling River, and the more fertile lands further south around the Murray Mouth, the Spencer Gulf, and along Bass Strait. They largely ignored the Eyre Peninsula, a small, lightly-settled agricultural land beyond the Spencer Gulf, since they deemed it too poor and too difficult to control without decent sailing technology. Direct imperial control did not always end with the desert; imperial forces maintained a few inland colonies to access some key resources such as the silver, zinc and lead of Broken Hill, and a few salt and gypsum harvesting colonies on some of the dry inland lakes.

In the inland regions of Australia, imperial influence was minimal, although they did have some contact with the desert peoples. Ancient trade routes stretched across much of the outback; ancient traders had travelled hundreds of kilometres across some desert routes when trading for flints and ochre. With the establishment of imperial outposts along the desert fringes, some of these trade routes were expanded. In a few locations with particularly high-value resources, the local hunter-gatherers found that they could mine a few key goods and trade these for food and metal tools from the agricultural peoples along the coast.

The most important of these routes became known as the Dog Road, which started at the imperial outpost of Port Augusta and ran over five hundred kilometres northwest to Coober Pedy. Here, the local Ngarjarli people mined opals from one of the richest sources in the world. The climate was far too dry to support a large population, or even a permanent population, but like most hunter-gatherer societies, the Ngarjali had a lot of under-used labour. Thanks to the imperial interest in opals, the Ngarjali found a reason to mine those gems and establish a semi-permanent settlement at Coober Pedy, where some of their people slowly mined opals throughout the year. Once a year, in June when the heat was least severe, the annual trading caravan set out from Port Augusta. People and dogs pulled travois loaded with trade goods: clay vessels full of wattle-seeds, smoked meats, dried fruits, and ganyu (yam wine); metal tools; textiles such as clothing, baskets and bags; and pituri (chewing tobacco). When they reached Coober Pedy, they held a great celebration and trading fair with the Ngarjali, and exchanged opals for their trade goods. This reliable food storage allowed the Ngarjali to occupy the same area for a large part of the year, although water shortages meant that they sometimes needed to move elsewhere.

* * *

While Watjubaga had clearly-defined natural borders in the north, west and south, its eastern frontier was more ambiguous. In most regions, imperial authority ran as far as the Great Dividing Range; the combination of rugged terrain, lack of beasts of burden, and distance from the imperial heartland meant that conquering the backward peoples of the eastern seaboard was usually not deemed to be worth the effort. In central Victoria, however, imperial armies had marched east from Junditmara lands and gained control of the lands around Port Philip Bay and West Gippsland. In the headwaters of the Murrumbidgee, the Monaro plateau was occupied by sullen imperial subjects who sometimes paid tribute and often rebelled. Further north, the rich farmlands of the Hunter Valley were inhabited by city-states who were reluctant imperial tributaries; this was the only region where imperial influence extended to the Tasman Sea. Apart from the Hunter, the eastern seaboard of Australia was independent of imperial control.

As an empire, Watjubaga thus claimed immense territory, but in many cases its level of control was limited. The empire maintained its predominance through its military strength, and more specifically through a core of well-equipped veteran soldiers who were the battlefield heavyweights of their day. In a land without cavalry, the imperial heavy infantry could be relied on to shatter any opposing army in any battle on open ground. Still, rebels often found ways to neutralise these tactics, particularly when fighting on irregular ground or resorting to raids and retreating to rugged terrain where the heavily-armoured imperial infantry had difficulty pursuing them.

Moreover, imperial manpower was limited. Watjubaga drew its soldiers exclusively from the ethnic Biral, who mostly dwelt in the ancient territories around Garrkimang, formed a significant minority in the rest of the Five Rivers, and elsewhere were either a small ruling elite or inhabited a few colonies established both as garrisons and trading posts. Joining the Biral was difficult for anyone of a foreign ethnicity; marrying in sometimes happened, but otherwise the only way to join was to persuade a Biral family to formally adopt someone, which was rare. These limitations on imperial manpower became an increasing strain with the large territories where Watjubaga tried to maintain its rule. The large distances and slow transportation technology meant that when away from one of the major rivers, even the local Biral elite often partially assimilated into the local culture, and the long lines of communication meant that local garrisons in distant territories were perpetually vulnerable to revolt.

* * *

At its height, Watjubaga ruled over vast territories, but it did not create much of a sense of unity amongst its subjects. With the Biral forming an elite ruling class, the subject peoples were not particularly inclined to adopt Biral language or culture. Some people learned Biral as a second language, since it was the language of government, but it did not become the primary language of any but the Biral themselves.

Still, while relatively few people could speak or write the Biral language, an increasing number of people in the Empire were literate. Later archaeologists would be aware of this by the wealth of written information preserved in clay tablets. Written accounts preserved considerable details about life within the empire, recorded in government records, legal documents and other archives, but also through an abundance of private documents such as letters, trade records, and religious texts. Within most regions of the Empire, government administrators could simply place tablets announcing new proclamations or other news in town squares, and be confident that they would be read, understood, and the information conveyed to everyone in the city.

The nature of imperial rule varied considerably amongst the imperial regions. Watjubaga’s core territories were the heavily-populated areas along the Murray and Murrumbidgee, and the almost as heavily-populated area around the Murray Mouth and along the Spencer Gulf. Here, imperial administrators exercised considerable control over everyday life, using a system of labour drafts which required every inhabitant to perform a certain number of days service for the government every year. This labour was required outside of the core harvest times, and was used to construct and repair public and religious buildings, maintain artificial wetlands, and sometimes to grow high-value crops such as corkwood (the key ingredient of pituri), which were subject to imperial monopoly.

Outside of the core territories, the labour draft system was much less prevalent. The imperial government tried to enforce it amongst the Junditmara in the south, with only limited success; this was one reason for the repeated revolts in that region. In the more thinly-inhabited regions along most of the Upper Darling, power was usually delegated to local chieftains instead. In the New England tablelands, the Empire ran the mines using a system of labour draft, but otherwise imperial control there was limited. In most other regions, the Empire did not even attempt to directly rule the territories, but simply collected tribute from local leaders.

In terms of religion, Watjubaga likewise exercised only limited control over the views of its subjects. The imperial view of religion was syncretic; like all of the Gunnagalic religions shaped during the Great Migrations, it had assimilated some indigenous beliefs from the hunter-gatherers displaced during the population movements. The underlying structure of their religion remained similar to their ancestors; they viewed the present world as only one aspect of the Evertime, the eternity which controlled everything and was everything. Within this framework, the actions of individual heroes, sacred places, and of spiritual beings were all adopted in a cheerful mishmash of beliefs. The Empire had no qualms about recognising other religious traditions as simply being aspects of the same underlying truth. Their only concern was for the religious role of the First Speakers, who had always maintained the claim that they were best suited to interpret the wisdom of eternity. Obedience to imperial authority was treated as accepting this religious duty; civil disobedience or outright rebellion were both treated as blasphemy. Beyond that, what individuals or people believed was of no concern to the imperial administration.

* * *

The Imperial era spanned several centuries, and it brought immense wealth to the royal city of Garrkimang. Extensive use of labour drafts usually meant that much of this wealth was invested into public architecture. The First Speakers and other noble classes amongst the Biral had a fondness for large, ornate buildings. At the height of imperial rule in 850 AD, Garrkimang had five separate palaces reserved for the royal family, and three dozen smaller palaces used by other noble families. They also built several large temples and many smaller shrines dedicated to various spiritual beings or former First Speakers. The royal city also held several large amphitheatres used for sporting and religious events.

Imperial engineering techniques were not particularly advanced by Old World standards, although they had developed considerably from their Classical ancestors. Imperial engineers built very effectively using a wide variety of stones. They had not discovered the arch, and lacked both the wheel and beasts of burden to help with moving building material around, but they had waterborne transport and lots of determination. Imperial construction techniques tended toward large, solid stone buildings, with the walls supported by buttresses at key points. They could build some very large columns, but they mostly used them for freestanding monuments or as aesthetic elements of building design, rather than as the main structural support. In the most elaborate imperial buildings, the solid buttressed walls were overhung with large eaves, and the eaves themselves were supported with elaborately-carved columns.

Imperial aesthetics placed great value on elaborate displays in architecture. This meant that imperial buildings were covered both within and without by a great many decorative elements: intricate ornamental stonework, sculptures, glazed tiles, murals, and above all bright, bright colours. Some valued stones are transported large distances because their appearance was preferred; the marble quarries at Bathurst and Orange were far from Garrkimang, but that was of little concern to the imperial engineers who order large quantities of marble to decorate the exterior of the palaces and temples. Colour was an integral part of most decorations, from some coloured stones, or glazed tiles, or from a variety of paints. While the individual stylistic elements were wholly alien to European building traditions, the overall impression of imperial architectural styles would be reminiscent of the Baroque period. In technicolour.

* * *

In technology, the advent of the Imperial era did not mark any dramatic improvement over the preceding Classical era. While the First Speakers were not hostile to new learning, the focus of imperial efforts was on administration, aesthetic improvements, and organisation, rather than any particular sense of innovation. Outside of engineering, architecture, and military technology, there were no fields where the First Speakers would be particularly interested in supporting experimentation or the application of new ideas.

Still, the spread of literacy allowed more communication of ideas, as did the growth of trade under the imperial peace. This contributed to some technological advances during the Imperial era. Metallurgy became considerably more advanced during this period, particularly in the development of many copper-based alloys. The exploration of the Broken Hill ore fields led to the isolation of zinc ores, and these were used to create brass. With imperial aesthetics being what they were, most brass and many alloys of copper with precious metals were used for decorative rather than functional purposes, although brass also came to be used in various musical instruments such as horns and bells. Imperial smiths knew of iron, both from ancient experience of meteoric iron, and as a waste product from their extraction of zinc ores [4]. However, their smelting techniques did not produce sufficient heat to melt iron ore, and so they did not make any significant use of the metal.

The spread of literacy allowed the beginnings of the development of a medical profession in the Empire. Doctors in the Imperial era began to make systematic studies of symptoms of sickness and injuries. Clay tablets found by later archaeologists included some handbooks of illnesses, of their diagnosis, prognosis and recommended treatments. Many of these recommended treatments did not actually work very well, since internal illnesses such as fevers, epilepsy and parasites were believed to be spiritual phenomena which required treatments by priests. Still, the early Imperial doctors had some capacity to assist in the treatment of physical injuries, using some basic surgical techniques, bandages, and a variety of lotions and herbal treatments derived from several plants to assist with treatment. They also had a basic knowledge of dentistry, using drills to deal with cavities, using forceps and other specialised tools to extract teeth, and using brass wires to stabilise broken jaws.

Imperial scholars had some knowledge of mathematics and astronomy, although their methods were often basic. They used some rudimentary trigonometry and related methods to assist with calculating engineering requirements, but they had little interest in algebra or other more advanced mathematical techniques. They kept astronomical records on matters which interested them, but they ignored some other aspects. They were aware of the movement of the planets, although they believed that both Venus and Mercury were each two separate bodies, not having made the connection between their appearances in the morning and evening. They kept enough of a watch over the constellations to recognise novas and supernovas. They kept particularly detailed records of comets, which they believed to be a visible representation of the reincarnation of a ‘great soul’ who would make their mark in the material world in the near future. Being born during the appearance of a comet was a highly auspicious omen, to the point where heirs to the imperial throne would sometimes be chosen based on their fact alone. They kept some occasional records of eclipses, although not systematically, and did not make any practical application of those records. Imperial scholars had no real conception of the shape of the earth; they still assumed that it was flat.

* * *

Militarily, Watjubaga reached its largest borders in 822 AD. One of their most celebrated generals, Weemiraga, had earlier subjugated the peoples around Port Philip Bay and West Gippsland, and incorporated them into the Empire. In 821-822, he made his great “March to the Sea,” leading an army across the Liverpool Range into the Hunter Valley, and then to the Tasman Sea. He imposed tributary status on the city-states in this region. This accomplishment would be recorded in sculptures, murals, andlegends, and it saw Weemiraga adopted into the royal family and become First Speaker from 838-853.

At this moment, it appeared that Watjubaga was in a period of ascendancy, but in fact these accomplishments were virtually the last military expansion which the Empire would achieve. Logistical difficulties meant that the Empire would find it difficult to expand further, and the imperial regime soon faced internal problems. After the death of Weemiraga, the succession was contested between three princes, leading to the worst civil war which the Empire had yet seen. The civil war ended in 858, but other underlying trends were further weakening imperial rule.

Imperial rule had relied on two pillars of the state: a solid core of veteran heavy infantry equipped with bronze weapons and armour from the continent’s main supply of the metal, who could defeat any major uprising, and a system of garrison-colonies in the far-flung regions of the Empire to maintain a local presence there and ensure the main army would be rarely needed. As the ninth century progressed, both of these pillars were being weakened. The colonisation of Tasmania in the early ninth century provided a rich new source of tin and bronze which was outside the imperial monopoly, disrupting the trade networks and government revenue, and allowing peoples in the southern regions of Australia to gain access to better arms and armour. Subject peoples were also becoming increasingly familiar with imperial military tactics, and through the legacy of several revolts developed ways to imitate or counter these military tactics. Combined with the increased bronze supply, this meant that rebellious peoples could now field soldiers to match the imperial heavy infantry, and the Watjubaga military advantage waned.

The other main pillar of the state was also being undermined by a slower but more gradual process of cultural assimilation. The Biral governors and upper classes formed a small minority in most of the outlying regions, and imperial governors came to look more to local interests than the dictates of a distant First Speaker in Garrkimang. Governors assumed more and more de facto independence, and successive First Speakers found it increasingly necessary to settle for payments of tribute and vague acknowledgement of imperial suzerainty, rather than maintaining any effective control.

The weakening of the imperial military advantage was manifested by two successive military disasters. In the 860s, imperial forces were sent to subjugate the Gippsland Lakes region in south-eastern Victoria. This was inhabited by the Kurnawal, a fiercely independent-minded people whose relatives had been one of the two main groups to cross Bass Strait and settle in Tasmania. Thanks to that colonisation, the mainland Kurnawal had access to good bronze weapons, and they largely fought off the imperial forces. The imperial commander conducted several raids and collected enough plunder to bring back to the First Speaker as a sign of victory, but the manifest truth was that the conquest had failed. Worse was to follow in 886, when a new campaign was intended to launch a new March to the Sea and conquer the Bungudjimay around Coffs Harbour. This time the imperial armies were defeated utterly, unable even to claim plunder. This marked the resurgence of the Bungudjimay as an independent people, and within the next few decades they would begin raids into imperial territory around the New England tablelands.

The defeat in Coffs Harbour marked a devastating blow to imperial prestige. Another disputed succession followed in the 890s. While this did not turn into a major civil war as had happened four decades earlier, it encouraged already-rebellious subject peoples. The Hunter Valley had always been a reluctant tributary, and actual payments of tribute had been largely non-existent since the 870s. In 899, the city-states of the Hunter ceased acknowledging even the pretence of imperial overlordship, and the weakened Empire was in no condition to restore its authority.

While the ruling classes in Garrkimang found it easy to disregard the loss of the Hunter tributaries, thinking of it as only a minor matter, a much more serious rebellion followed. The Junditmara peoples had long resented foreign rule, requiring substantial imperial garrisons. A revolt over labour drafts in 905 provided a trigger for unrest, and in the next year it turned into a general Junditmara revolt. The imperial troops were massacred or driven out of Junditmara-inhabited territory, and in 907 the army sent to reconquer them was outnumbered and defeated. The Junditmara peoples established their own loose confederation to replace imperial rule. They would take what they had learned of imperial technology, literacy, astronomy and other knowledge, and apply it to their own ends. The loss of the Junditmara lands also made imperial rule over the rest of southern Victoria untenable, and they lost everything south of the Great Dividing Range within a few years of the Junditmara establishing independence.

With crumbling imperial authority in the south, the First Speakers turned to one last territorial expansion. The Eyre Peninsula, beyond the Spencer Gulf, had long been disregarded by the Empire. The peninsula was a small region of fertile land separated by a desert barrier from the nearest imperial city at Port Augusta. The land was useful for agriculture, and had some very occasional trading links with the Yuduwungu across the western deserts, but had otherwise not much to recommend it, and the separation of deserts and water made a military campaign difficult. Keen to restore some military prestige, the imperial government cared little for such details, and despatched forces who marched overland from Port Augusta. The Peninsula peoples withdrew behind city walls, and although these were besieged and captured one by one, the long and bloody warfare did not justify the conquest. While the Eyre Peninsula was proclaimed as conquered in 926, the loss of imperial manpower would hurt far more than the minor gain in resources.

After the conquest of the Eyre Peninsula, the remaining imperial structures started to rot from the periphery inward. In the north, the local governors assumed effective independence, although the fiction of imperial control continued for two more decades. The decisive break came in 945. The governors of the five provinces which made up the region of New England had long been more sympathetic to their subjects than the distant proclamations of the First Speakers. The governors announced their secession from the Empire in a joint declaration in 945, bringing the local Biral garrison-colonies with them, and raising additional local forces for defence. Imperial forces were sent to reassert control of the source of tin (and many gems), and were defeated in a series of battles in 946-948. This event, more than anything else, marked the collapse of the Empire. Apart from some brief attempts to reconquer a rebellious Eyre Peninsula in the 970s, this marked the last time that the Empire would try to project military power outside of the heartland of the Five Rivers.

Imperial rule over the core of the Murray basin – the Five Rivers – persisted for much longer than the more distant territories, but with the same condition of gradual decline of imperial power. The Biral remained a resented ruling class along the Murray proper, and revolts became increasingly common. The First Speakers resorted to increasingly desperate measures to quell some revolts, including the wholesale razing of Weenaratta in 1043, but in the end, none of these measures were successful. The lands around the Murray Mouth were lost in the 1020s, Tjibarr rebelled in 1057 and started to encroach further into imperial territory, and Gutjanal [Albury-Wodonga] asserted its independence in 1071, taking most of the dominions of the old kingdom of Gundabingee with it. By 1080, the Empire consisted of little more than Garrkimang and its immediate hinterland; its borders had shrunk even further than the borders of the Classical kingdom. Internal revolt removed the last of the First Speakers in 1124, leaving Garrkimang a decaying city filled with monuments to past imperial glories.

* * *

[1] The Kangaroos came from Gray, the Corellas from White, the Ravens from Black, the Kookaburras from Blue, the Echidnas from Azure, and the Possums from Red. The old colours of Gold and Green were lost during the migrations.

[2] In modern Australia, this region has been transformed into the Murrumbidgee Irrigation Area. This uses a system of weirs, canals and holding ponds to irrigate the area, and which is in turn fed by larger dams further upriver. This has made the area very productive agriculturally. In allohistorical Australia, Garrkimang engineers have developed their own complex system of dams and weirs to trap floodwaters and feed them into wetlands which are used, as elsewhere, for fishing and hunting.

[3] The Five Rivers are the Murray, the Murrumbidgee, the Lachlan, the Macquarie, and the Darling, all of which are part of the Murray-Darling basin. The earlier Three Rivers did not include the Murray and Darling Rivers, but the nation was renamed when it extended control over those two major rivers.

[4] The process which the Imperial smiths have used to develop brass is distinct from that used elsewhere in the world. Early brasses elsewhere were produced from calamine, which is an ore which contains zinc carbonate and zinc silicates, and which were melted with copper to produce brass. In *Australia, the Imperial smiths have explored the massive Broken Hill ore deposits, initially for extraction of lead and native silver, both of which are abundant there. Mining in this deposit will also mean that they discover sphalerite, an ore of zinc sulfide which has also has impurities of iron. This can be melted to produce brass in the same way that calamine was elsewhere, but it also means that iron will frequently be encountered as an impurity in the waste products.

* * *

Thoughts?
 
So the Watjubaga are somewhat of a romanesque empire? I like it, but since I've never been to Australia or really gotten into where everything is, a map would be extremely helpful.

Also, did you mention a system of roads that the Watjubaga created? Or are they more decntralized than that?
 

mojojojo

Gone Fishin'
We have had maps of this TL. I would love to see, some artist take a crack at the buildings and peoples!
 
I'd bet the Watjubaga are closer to the imperial powers of the ancient Eastern Mediterranean or the Fertile Crescent than the Romans.

Their heavy infantry sounds a lot like the hoplites.
 
Brilliant post Jared, I do love the whole grand sweep of history thing this TL is going with so far. Looking forward to the post-Imperial era!


So the Watjubaga are somewhat of a romanesque empire? I like it, but since I've never been to Australia or really gotten into where everything is, a map would be extremely helpful.

I'm working on that! Hopefully Jared will be able to post something soon.
 
Excellent update! I like how you described the rise and fall of the Watjubaga, it sounds a lot like Australia's Roman Empire.

EdT: Can't wait to see your map. :)
 
Jared's asked me to put this up, so...

Red-and-Gold1.png
 
Fascinating history, and wonderful map.

I do have one question, though. Given the rudimentary-at-best naval skills of the Australians, how exactly does trade with and migration into Tasmania work?
 

Fatal Wit

Banned
In culture and religion, Garrkimang was also unlike any other Classical city. The inhabitants called themselves the Biral, a name which means roughly “chosen ones.” They traced this back to the migration under Wunirugal, believing that they had been chosen to be granted their new land as a sacred trust. Their religion had a similar foundation to the older Gunnagal beliefs; they still shared the same general view of the Evertime and of the spirit-beings who inhabit eternity, although they gave different names and attributes to many of those beings. Yet the old beliefs had been overlaid by a new religious structure, that of the First Speakers and their representatives who interpreted the world.
Interesting.... and it wouldn't surprise me if the Garrkimang end up being the target of assertions of an ancient Hebrew presence in Australia(much like "historical" revisionists made such claims about the Americas). After all, "chosen ones" granted a sacred plot of land:p... it would be cool of this alternate Australia had the equivalent of the Mormons making similar claims about Jesus having visited Australia and left a sacred revelation.

Oh, and fantastic maps- they look professional, to say the least.
 

Riain

Banned
I went for a good drive around this TLs Junditmara area named Tuhonong on the map on the weekend and learned two things.

1) You can't get batteries for a digital camera in the middle of nowhere on Easter Saturday morning.

2) TTLs Tuhonong is between two volcanos, Mt Eccles and Mt Napier, thus there is plenty of loose stone all around the swamp to make the wiers and dams to improve the wetland. Is this convenient building material a major factor in the OTL Condah swamp being a improved wetland with a substantial resident population? So even if the idea spread beyond Condah (Tuhonong) could the lack of convenient building materials, especially with the rudimentary means of moving these building materials, prevent otherwise suitable wetland from being improved? I've never really spent a lot of time up in the Murray-Darling basin so don't know if there is plenty of loose stone handy to build weirs and dams in wetlands.
 
Jared, can I just say for one that if you cause an outbreak of Marnitja in Europe during the Thirty Years' War, I will personally worship you as a god. Because any TL where you can kill that damned Swede Gustav Adolph with a disease that eats your lungs...priceless.

Oh, Marnitja will definitely hit Europe before the Thirty Years War is out. It's more or less inevitable, since this is a disease which produces a significant number of asymptomatic carriers. (As do the real equivalents.) If there are, say, five hundred Dutch *conquistadors sent to invade Australia, and if they catch Marnitja while they're there, then somewhere between 2-8 of them will become asymptomatic carriers. All it takes is at least one of those returning home to Europe, and there's a pandemic coming.

The main question is when, exactly, it hits. I'd say sometime in the 1620s, or early 1630s at the latest, depending on how many people go back and forth. I need to look into the population numbers a bit for that. When it does, ouch, ouch, ouch. I'm thinking of starting a plannification thread for the global effects of Dutch contact with *Australia, and this will be one of the major areas covered. Although that can wait until I've shown more of what Australia itself is like in the pre-1619 days.

In terms of individual historical figures, by the way, I'm relying on an unbiased method of working out who survives, and in what condition they survive. Gustavus Adolphus is only one person I'll need to figure out. Life gets very butterflied if Cardinal Richelieu becomes red inside and out.

And I guess cannibalistic orgies for the Atjuntja temples, out of sheer weirdness? Do I get a cookie?

Not quite what they're up to, although the effects are equally unpleasant.

So the Watjubaga are somewhat of a romanesque empire? I like it, but since I've never been to Australia or really gotten into where everything is, a map would be extremely helpful.

Thanks go to EdT for the map. The Watjubaga are a composite of various real empires, plus a few bits which were inspired by some of the earlier aspects of the Gunnaglic cultural development. There were a couple of Roman-style elements in their attitude toward religion, although that's partially based on the Mongol approach, too, and their use of garrison-colonies, although that was also partly based on the system which the early Assyrians used. The use of public pronouncements and law codes in town squares was something which the Romans did, but it also wasn't exclusive to them - the Babylonians did it, too.

Also, did you mention a system of roads that the Watjubaga created? Or are they more decntralized than that?

They have a few major roads here and there, similar to what various Middle Eastern empires did - the Royal Roads of pre-Hellenic Persia, for instance. But they're nowhere near as organised about it as the Romans were. Wherever possible, they prefer to move things by water than by road, which is one reason they had trouble holding onto areas where water transport wasn't viable.

We have had maps of this TL. I would love to see, some artist take a crack at the buildings and peoples!

I'd like that too, but my artistic ability to show people isn't much above the stick-figure level.

I'd bet the Watjubaga are closer to the imperial powers of the ancient Eastern Mediterranean or the Fertile Crescent than the Romans.

Their heavy infantry sounds a lot like the hoplites.

In general, they're a combination of various Eastern Mediterranean imperial powers with a few other attributes thrown in. Their military structure is something like citizen-soldier hoplites such as the Greek used, although their use of skirmishers is quite unlike the Greek attitude to them. Their institution of an absolute monarchy is pretty much lifted straight from any of the absolute monarchs of Assyria or Babylon. Their military expansion also has a lot of similarities to the Assyrian model, and their great general Weemiraga was partly inspired by the Assyrian monarch Tiglath-Pileser I. Their inclusion of an ethnic elite which it is hard to join, but where they set up colonies, is much like the early Assyrians were - the early Assyrians had merchant colonies called karum which are much like those that the Biral/Watjubaga use. Their views on religion were mostly inspired by the Mongol attitudes (with a dash of Roman), where obedience is treated as religious submission, too.

Brilliant post Jared, I do love the whole grand sweep of history thing this TL is going with so far. Looking forward to the post-Imperial era!

The post-Imperial era is going to be shown in more complex ways. The next post shows more about first contact between the Maori and the Raduru, the Gunnagalic people who live in the Illawarra region of New South Wales. The post after that is an overview post which gives a quick catch-up on history up until 1618, and then shows what Australasia is like in 1618. After that, it's a case of showcasing some individual cultures in more detail, starting with the Atjuntja. As part of that, it will show what the various cultures were like during the post-Imperial era (which extends to 1618, of course).

Excellent update! I like how you described the rise and fall of the Watjubaga, it sounds a lot like Australia's Roman Empire

Tis an awesome and sweeping update Jared. Kudos to you. :cool:

WOW:cool::cool: Cool update, I also love the map.

Glad y'all like it.

Fascinating history, and wonderful map.

I do have one question, though. Given the rudimentary-at-best naval skills of the Australians, how exactly does trade with and migration into Tasmania work?

Their naval skills and shipbulding technology isn't completely static; they are slowly becoming better at building ships. Mostly for fishing purposes, and Bass Strait does offer some pretty good fishing. OTL Aboriginal peoples used to do a surprising amount of fishing in Bass Strait even when all they had was bark-skin canoes; I read an amusing account of a European ship, the Marie Gabrielle, which was having trouble going around Cape Otway in rough seas... and when it did come past Cape Otway, it found the local Aboriginal people happily fishing in the open sea from those bark-skin canoes. The boats which the coastal Aboriginal peoples have in *Australia are better than that, if not up to Greco-Roman standards. It's possible to sail back and forth between Tasmania and the mainland once someone knows that it's there.

The actual discovery of Tasmania is an accident, though, and not done in one go. What happens is that boats do sometimes get blown out to sea, and sometimes they are able to make their way home again. They discovered some of the Bass Strait islands. They settle King Island and Flinders Island first (two separate groups of people), which were uninhabited. From here, they become aware of Tasmania... and they soon discover what Tasmania holds: tin. Large reserves of tin. Tin was the driver of long-distance trade in the Bronze Age of the old world; it was traded from Cornwall at least as far as Italy. A similar inspiration brings people to Tasmania. That provides the impetus for colonisation, and the profit which comes from shipping tin back to the mainland is well worth the risks which people take sailing across Bass Strait. (And it is a risk; quite a few boats don't make it.) There aren't actually that many other goods which are considered worth trading; gems, mostly, and a few other luxuries. But tin is the main driver of trade and continued contact with the mainland.

Interesting.... and it wouldn't surprise me if the Garrkimang end up being the target of assertions of an ancient Hebrew presence in Australia(much like "historical" revisionists made such claims about the Americas). After all, "chosen ones" granted a sacred plot of land:p... it would be cool of this alternate Australia had the equivalent of the Mormons making similar claims about Jesus having visited Australia and left a sacred revelation.

There could be all sorts of interesting theories, in the long-term. Often entertaining, and completely incorrect, of course, but that doesn't stop people speculating. Including about the possibility of pre-Houtman contact with Australia...

I went for a good drive around this TLs Junditmara area named Tuhonong on the map on the weekend and learned two things.

1) You can't get batteries for a digital camera in the middle of nowhere on Easter Saturday morning.

2) TTLs Tuhonong is between two volcanos, Mt Eccles and Mt Napier, thus there is plenty of loose stone all around the swamp to make the wiers and dams to improve the wetland. Is this convenient building material a major factor in the OTL Condah swamp being a improved wetland with a substantial resident population? So even if the idea spread beyond Condah (Tuhonong) could the lack of convenient building materials, especially with the rudimentary means of moving these building materials, prevent otherwise suitable wetland from being improved? I've never really spent a lot of time up in the Murray-Darling basin so don't know if there is plenty of loose stone handy to build weirs and dams in wetlands.

In most places, there's enough stone around for the sorts of weirs and dams which they'd be building. Most of this is sedimentary rock - essentially, the Murray basin is covered in sedimentary layers, which are pretty easy to work. And fortunately they can move things along the main rivers, which makes it easier to bring it in even to those places where there's local stone lacking. Trade by water along the Murray happened even in pre-agricultural peoples, so I suspect that they could import enough building stone if they needed it.

Edit: Tuhonong isn't exactly Lake Condah, by the way - it's in the same location as Hamilton, Victoria, which is a little north of Lake Condah.
 
I do hope the Dutch aren't going to pwn the Gunnagal, et al, in the same way the Spanish pwned the South and Central Americans.
 
Jared's asked me to put this up, so...
Great map, EdT!:)

Still, while relatively few people could speak or write the Biral language, an increasing number of people in the Empire were literate. Later archaeologists would be aware of this by the wealth of written information preserved in clay tablets. Written accounts preserved considerable details about life within the empire, recorded in government records, legal documents and other archives, but also through an abundance of private documents such as letters, trade records, and religious texts. Within most regions of the Empire, government administrators could simply place tablets announcing new proclamations or other news in town squares, and be confident that they would be read, understood, and the information conveyed to everyone in the city.

...

* * *

In technology, the advent of the Imperial era did not mark any dramatic improvement over the preceding Classical era. While the First Speakers were not hostile to new learning, the focus of imperial efforts was on administration, aesthetic improvements, and organisation, rather than any particular sense of innovation. Outside of engineering, architecture, and military technology, there were no fields where the First Speakers would be particularly interested in supporting experimentation or the application of new ideas.

Still, the spread of literacy allowed more communication of ideas, as did the growth of trade under the imperial peace. This contributed to some technological advances during the Imperial era. Metallurgy became considerably more advanced during this period, particularly in the development of many copper-based alloys. The exploration of the Broken Hill ore fields led to the isolation of zinc ores, and these were used to create brass. With imperial aesthetics being what they were, most brass and many alloys of copper with precious metals were used for decorative rather than functional purposes, although brass also came to be used in various musical instruments such as horns and bells. Imperial smiths knew of iron, both from ancient experience of meteoric iron, and as a waste product from their extraction of zinc ores [4]. However, their smelting techniques did not produce sufficient heat to melt iron ore, and so they did not make any significant use of the metal.

The spread of literacy allowed the beginnings of the development of a medical profession in the Empire. Doctors in the Imperial era began to make systematic studies of symptoms of sickness and injuries. Clay tablets found by later archaeologists included some handbooks of illnesses, of their diagnosis, prognosis and recommended treatments. Many of these recommended treatments did not actually work very well, since internal illnesses such as fevers, epilepsy and parasites were believed to be spiritual phenomena which required treatments by priests. Still, the early Imperial doctors had some capacity to assist in the treatment of physical injuries, using some basic surgical techniques, bandages, and a variety of lotions and herbal treatments derived from several plants to assist with treatment. They also had a basic knowledge of dentistry, using drills to deal with cavities, using forceps and other specialised tools to extract teeth, and using brass wires to stabilise broken jaws.

Imperial scholars had some knowledge of mathematics and astronomy, although their methods were often basic. They used some rudimentary trigonometry and related methods to assist with calculating engineering requirements, but they had little interest in algebra or other more advanced mathematical techniques. They kept astronomical records on matters which interested them, but they ignored some other aspects. They were aware of the movement of the planets, although they believed that both Venus and Mercury were each two separate bodies, not having made the connection between their appearances in the morning and evening. They kept enough of a watch over the constellations to recognise novas and supernovas. They kept particularly detailed records of comets, which they believed to be a visible representation of the reincarnation of a ‘great soul’ who would make their mark in the material world in the near future. Being born during the appearance of a comet was a highly auspicious omen, to the point where heirs to the imperial throne would sometimes be chosen based on their fact alone. They kept some occasional records of eclipses, although not systematically, and did not make any practical application of those records. Imperial scholars had no real conception of the shape of the earth; they still assumed that it was flat.

* * *
...
While the ruling classes in Garrkimang found it easy to disregard the loss of the Hunter tributaries, thinking of it as only a minor matter, a much more serious rebellion followed. The Junditmara peoples had long resented foreign rule, requiring substantial imperial garrisons. A revolt over labour drafts in 905 provided a trigger for unrest, and in the next year it turned into a general Junditmara revolt. The imperial troops were massacred or driven out of Junditmara-inhabited territory, and in 907 the army sent to reconquer them was outnumbered and defeated. The Junditmara peoples established their own loose confederation to replace imperial rule. They would take what they had learned of imperial technology, literacy, astronomy and other knowledge, and apply it to their own ends.
The location of Empire may mean that Europeans won't see the more developed parts of *Australia until a certain time later, unless the area suffers a further decline, but the spread of education may delay or prevent that.
 
Last edited:
Status
Not open for further replies.
Top