"Peak Oil" is not such a strange or unique phenomenon. It is, essentially, just another case of diminishing returns. Hubbert's Peak is the point of diminishing returns, the midpoint in global oil production where half the oil is consumed, and half remains. Of course, the half that was taken first was the half that was easiest to extract and had the highest quality. It was light, sweet crude bubbling near the surface. What remains is heavy, sour crude deep inside the earth, or in forms that are otherwise difficult to process, such as tar sands and shale. These are offered up as hope that Peak Oil may not be so bad, but they miss the point entirely. The importance of oil is not that it provides energy; energy can be had from anything. The importance of oil is that is provides cheap energy. A society's complexity is not a function of the total energy throughput, but the ERoEI--Enery Returned on Energy Invested, or ROI in pure energy economics terms. Since the general problem (if not the specifics) is such a common one, allow me to explain with an example from our own history: the end of the Bronze Age, the beginning of the Iron Age, and a crisis we might today call, "Peak Wood."

We can think of metal tools in terms of ERoEI, as well. A certain amount of energy goes into their creation--the energy of the smith, the energy of the miner, the energy of the fuel used for smelting, and the energy used in transporting all those elements to one another. There is a return on that investment, as well--usually in the form of the wedge. What constrained the proliferation of the Bronze Age, however, was lack of fuel, just like our current crisis. Richard Cowen describes the situation well in his essay on the Bronze Age:

Egypt, which has practically no trees, was trading with Byblos (on the Lebanese coast) for cedar for shipbuilding, temple construction, and furniture-making as early as 3000 BC. But perhaps the most famous documentation of the shortage of wood around the ancient Mediterranean is the Epic of Gilgamesh ... Stripped of sex and violence, the Gilgamesh epic is about deforestation. Gilgamesh and his companion go off to cut down a cedar forest, braving the wrath of the forest god Humbaba, who has been entrusted with forest conservation. It's interesting that Gilgamesh is cast as the hero, even though he has the typical logger mentality: cut it down, and never mind the consequences. The repercussions for Gilgamesh are severe: he loses his chance of immortality, for example. But the consequences for Sumeria were even worse. It's clear that the geography and climate of southern Mesopotamia would not provide the wood fuel to support a Bronze Age civilization that worked metal, built large cities, and constructed canals and ceremonial centers that used wood, plaster, and bricks. Most timber would have to be imported from the surrounding mountains, and deforestation there, in a climate that receives occasional torrential storms, would have led to severe erosion and run-off. The loss of Gilgamesh's immortality may be a literary reflection of the realization that Sumeria could not be sustained.

Theodore Wertime suggested that massive deforestation of the eastern Mediterranean began about 1200 BC, for construction, lime kilning, and ore smelting. Probably it began earlier in the drier regions further east. King Hammurabi's laws (around 1750 BC) carried the death penalty for unauthorized felling of trees in Mesopotamia. The problem may have been even worse in intensive metal-working regions like Anatolia. Metal smelting and forging had been going on in Anatolia for at least 3000 years by 1200 BC.

...

Copper smelting needs a great deal of fuel, especially if the ore supply is dominantly sulfide. About 300 kg of charcoal are needed to produce 1 kg of copper by smelting 30 kg of sulfide ore. A tonne of charcoal needs somewhere between 12 and 20 cubic meters of wood, and for each cubic meter of wood a 100-year old tree has to be felled.

Archaeologists have estimated that the Bronze Age copper mines at Mitterberg, in the Austrian Tyrol near Salzburg, must have employed about 180 miners and smelters to produce about 20 tonnes of copper a year. Then one has to add the woodcutters, carpenters, charcoal burners, and carters who cut, carried, and processed the wood needed for the gallery timbers, the firing of the working face, and the fuel for the furnaces, and then add the farmers that fed all these. This was a very large-scale operation.

In copper smelting we find, perhaps, the first major environmental effect of mining. The Mitterberg copper mine probably required about 19 acres of forest to be felled each year, just for the smelters. Even with efficient natural regeneration of the forest, this is a sustainable harvest from perhaps 2 square miles of forest. In fact, however, the cleared land was probably used not for re-growth but at least partly for agriculture, to support the mining community.

On a time scale longer than 10 years, however, a Bronze Age copper mining operation must have caused local deforestation on a large scale, and ever-increasing costs for hauling the wood to keep the industry going.

...

The tremendous tonnage of ancient copper slag on Cyprus suggests that the Cypriot copper industry collapsed around 300 AD simply because the island ran out of cheap fuel. The slagheaps suggest a total production of perhaps 200,000 tonnes of copper, and that in turn suggests that fuel equivalent to 200 million pine trees were cut to supply the copper industry, forests 16 times the total area of the island. Even given that high-altitude Cypriot forests can regenerate quickly in the right conditions, this suggests that wood fuel was a critical constraining factor on the Cypriot copper industry, and must have been a persistent problem on the island for other industries too.

The landscape of Cyprus today (and Greece, and Turkey, and Lebanon, and in fact most of the Mediterranean seaboard) is quite unlike its appearance 5000 years ago. The magnificent cedar forests of Lebanon were felled largely for timber for buildings and ships, but copper smelting must take most of the blame in Cyprus. This Mediterranean ecological disaster used to be blamed on the Arab introduction of goats to the region several centuries AD, but the change was much earlier. There are secondary effects of deforestation, of course: hillsides are exposed to greater run-off, and erosion can be greatly accelerated. Part of the story of the later Bronze Age seems to be the silting of coastal ports and cities. The city of Tiryns, for example, spent a great deal of effort just before its end building a diversion structure to keep floods out.

The timber crisis of the late Bronze Age was obviously not the extinction of all trees in the world. It didn't need to be, just as we don't need to run out of oil to face a similar fuel crisis. There was still lumber to be felled; but as Bronze Age kingdoms deforested their surrounding ecosystem, the nearest forest became farther and father away. Loggers had to travel farther to reach the forest, and once the trees were felled, they needed to be transported longer and longer distances back home. The energy invested was constantly increasing, but the energy returned remained the same. The ERoEI plummeted, and Bronze Age civilization collapsed into a dark age for several centuries. Civilization finally re-emerged with the Iron Age. Iron had been used throughout the Bronze Age as well, but primarily for ceremonial artifacts. It was expensive. Iron took greater amounts of labor, and for that cost derived an inferior product. Iron could not keep its edge as effectively as bronze; it was heavier and more brittle. Additionally, it was ugly; it was for this that Hesiod wrote, "And I wish that I were not part of [this] generation of men, but had died before it came, or been born afterward. For here now is the age of iron."

As the ERoEI of timber plummeted, so, too, did the ERoEI of bronze. Eventually, despite its higher cost and lower quality, the ERoEI of bronze dropped so low that even iron was better. Clayton Cramer writes in "What Caused the Iron Age?" [PDF]:

It is also apparent that ferrous metals displaced bronze in tools before weapons and armor. A tool that loses its edge or bends prematurely is a nuisance; a sword that loses its edge in battle puts a warrior’s life at risk. This tends to confirm the theory that iron was originally a second-best alternative to bronze, until the accidental discovery of carburization.

What ushered in the Iron Age was the discovery of carburization--the process of working and reworking iron in charcoal, allowing the iron to alloy with carbon, creating steel. Cramer writes again:

Whether the underlying cause was a tin shortage, or a copper shortage, it is easy to understand why Eastern Mediterranean societies first turned to iron as a cheaper, less effective alternative to bronze. After the discovery of carburization and quenching, steel was both cheaper and more effective than bronze.

Steel was a deus ex machina that allowed civilization to continue its rampant expansion and exploitation. In other scenarios, however, the "Peak Wood" problem was fatal. It was the proximate cause for the collapse of both Cahokia and the Hohokam. It nearly did in European civilization again in the 16th century--that time, the deus ex machina was coal.

Every civilization eventually falls prey to diminishing returns. The problem of Peak Oil--like "Peak Wood"--is just one dimension of this much larger, intractable problem, inherent to the nature of any complex society. What separates extant civilizations from extinct ones is whether or not a less attractive alternative existed, which could become the basic strategy for a new iteration in the cycle of expansion and exploitation. But eventually, miracles run out. Eventually, the deus ex machina leaves us to sink or swim on our own merit. The crisis of Peak Oil is precisely the kind of crisis that has always collapsed civilizations, and if history is any guide, then it seems very likely that we have finally run out of luck, and the time has finally come to pay back 10,000 years of debt.