I would caution you about making such broad suppositions as there will be no more arable land, or that “agriculture is simply no longer possible without petroleum, except in very rare exceptions.”
It would be more useful to discuss particular geographical areas, rather than making such blanket statements.
I disagree that agriculture is necessarily unsustainable. It certainly can be and is, especially when run by petroleum, but I think that a balance of horticulture, foraging, and low-impact non-industrialized farming and patoralism (in the rhizome scheme of things) can be sustainable, in a variegated manner in differing geographical locations.

In my, admittedly limited, experience with forest gardens, horticulture, and industrial-scale (organic as well as conventional) farming, I don’t see how agriculture could be sustainable in the true sense of the word. Agriculture tries to constantly reproduce the conditions of a limited, very specific climatic area, like that of the Fertile Crescent when it still lived up to its name. It does not take into account particular geographical areas–quite the contrary, actually. One of its staple crops are grains, grasses, which in nature seem to serve as an intermediate, undemanding ground cover to protect the soil from erosion, permit life to loosen and enrichen the soil underneath this protective blanket, and provide a habitat for low-growing plants in contrast to neighbouring areas with different types of vegetation. The grasses eventually yield to bushes and then trees, if the climate permits.

In agriculture, the vast plain areas of grasses, called fields and set up to mimick the original conditions of the Fertile Crescent, are not succeeded by more complex forms of vegetation typical of that particular geographical area: The grass itself produces the desired harvest. Humans go to great lenghts to ensure that new grass can be grown and harvested year after year after year. They labouriously pull out plants that might outcompete their precious grass, cultivate the land to vastly reduce the chances of anything but grasses to grow, and, lately, poison everything except for, and often including, their grass. After harvest, the fields lay bare, although the grasses had been there to keep exactly this from occuring in the first place.

It seems as if the trauma of being driven out of the Garden of Eden has made us cling to the last straw, that of the grain, to keep us from having to face Death. This desperate, rigorous grip manifested itself in agriculture, which continues to shape every aspect of human life, and all the other forms of life as well. Paradoxically, it appears to have brought us even closer to extinction than we were, back when the last straw offered itself to us. Also, our agriculturally-produced food is not exactly healthy to us, and the aspect of addictive qualities, especially of wheat, adds to this.

Perhaps we can only free ourselves from this paradox by loosening our grip of the grain and consciously taking the plunge into the undiscovered country which lies beyond. Either that, or climate change will do it, as the era of stable weather comes to an end. Or any number of other events will do it.

The main point I’m trying to make is that Collapse will force humans to creatively interact with their (changing) environments and express their adaptability based upon their local ecosystems.
As there is a broad diversity of ecosystems, and humans, I think that the expressions of human adaptability will be equally diverse. Across the social spectrum this will include foragers, agrarian folks, and raiders, and many others we haven’t even thought of yet.

Of course, and this is what makes it worthwhile.

Yet, the story does not end with the physical aspect of agriculture’s problems, and this is where the primitivist worldview comes into play: In the context of a complex (not just complicated) world, within the boundaries set by the capabilities and vulnerabilities of humans, agriculture, and perhaps even horticulture, seem to invariably lead to a surplus which has to be dealt with. Just to offset the effects of the growing seasons, we have to store the harvest until the next one can be reaped.

And as the harvest represents pure energy, it brings power to whoever can access it, power which has to be counter-balanced, then regained, and so on, leaving us in a constant rush to fix ever-increasing numbers of unforeseen side effects of previous fixes and power struggles. In the process, as long as there is the energy to fuel it, complexity builds on all levels, overshadowing those aspects of life which take time (or now-time), thorough reflection, continuous observation of our world, in other words: the opposite of the constant rush.

If it is our aim to return to the Garden, then we will have to find ways to keep us from using accumulated energy of any kind. If this means communal self-restriction to life as a hunter-gatherer, then, a primitivist would say, then be it! Although there may be other ways, ways yet to be discovered, maybe not by our species, but in due time.

Again, what I’m saying is, the unknown changes brought about by Collapse will be regionalized, as will the response and adaptation to them, as opposed to any uniform “results.”

I believe that this is more or less exactly what Jason wrote about in the Spirit of Place article.

(P.S.: Sorry if this comment comes out as badly formatted, I’m not sure if the blockquote element will work as expected)

What about lands that get their fertility of rivers, like the Nile - it can support a pretty complex society.

Not since they built the Aswan dam. That’s caused a great deal of damage to the Nile’s agricultural capacity. In the long run, as argued in thesis #29, global climate change will end the Holocene, and with it, the age of agriculture. In the short term, soil exhaustion eliminates the vast majority of our possibilities. The Nile might become a possibility if the Aswan dam is destroyed, but only in the short term. Meanwhile, the Fertile Crescent was exhausted millennia ago. These will only be able to support localized pockets of agriculture. If you’re caught in Egypt’s surviving civilization and want to get out a century from now, a few miles into the desert will be all it takes to be far beyond their reach.

I remember the black fertile soil of Ukraine where I was born. Europian lands might have been depleted faster after the green revolution, but there must be still much nutrients to support agriculture.

Actually, it’s mostly the exhaustion that led to the Green Revolution that concerns me. Perhaps Ukraine is exceptional, and will retain one of those possible pockets, I don’t really have the data to say. But the Green Revolution was not adopted without reason. Agriculture has been a 10,000 year race against its own consequences. For most of that time, the strategy was to gobble up more land, before the old land gave out. Usually, new land was acquired just in time. The colonization of the New World began in earnest only when the crops of the Old World began to fail, for instance. That strategy ended in 1960, when we finished farming the last bits of arable land left. After that, we turned to the Green Revolution, and leveraging other ecological resources in more complex ways, to continue farming the same land. So in effect, the expansion continues, but we’ve found ways to farm the whole biosphere now. When complexity breaks down, though, this will stop, and we’ll need to face the consequences of agriculture: namely, the exhaustion of all the soil suitable for agriculture. There will be pockets and exceptions, yes, but they will be pockets and exceptions. Most of the land that is currently farmed could not be farmed without tractors, petrochemical fertilizers, and other complex artifacts of industrial society. When complexity breaks down, so will the ability to farm most of our cropland.

What about slash-and-burn techniques that depend on forests rather than fertile land? I don’t see it all gone in the next thousands of years.

No, that’s a form of horticulture, rather than agriculture, and is thus subject to an entirely different dynamic. Many types of horticulture might have a long future ahead of them.

It would be more useful to discuss particular geographical areas, rather than making such blanket statements.

We’ve been doing some of that. Right now, we’ve got two entries in our “Shape of Collapse” series: China and India and Small Town America.

I disagree that agriculture is necessarily unsustainable. It certainly can be and is, especially when run by petroleum, but I think that a balance of horticulture, foraging, and low-impact non-industrialized farming and patoralism (in the rhizome scheme of things) can be sustainable, in a variegated manner in differing geographical locations.

Firstly, horticulture and agriculture are very different things. See thesis #8, or Toby Hemenway’s latest, “Is Sustainable Agriculture an Oxymoron?”

Secondly, agriculture was causing widespread devastation long before the Green Revolution. The Fertile Crescent was, once upon a time, a very fertile region. It was turned into a desert by agriculture. Richard Manning’s Against the Grain is an excellent history of agriculture’s horrible consequences, even before the Green Revolution; his classic Harper’s article “The Oil We Eat” presents much of that material in a more condensed form; or, you can simply listen to our interview with him in episode #3 of our podcast.

The main point I’m trying to make is that Collapse will force humans to creatively interact with their (changing) environments and express their adaptability based upon their local ecosystems.

Absolutely, but agricultural peoples the world over are remarkably homogeneous. The agricultural way of life is so difficult, dangerous and unhealthy that there are very few ways to make it work wherein those participating in it even survive. Thus, pretty much all of humanity’s cultural diversity—and it is an enormous range—is made up of those peoples we chauvanistically lump under the heading of “foragers.” This encompasses everything from the Inuit to the !Kung to the Plains Indians to the Pygmies. Collapse will be met with enormous creativity (we’re trying to get some of that creativity flowing with The Fifth World), but it’s also a collapse, and that means certain things happen. The physical reality changes, and some things that used to be impossible become possible, while some things that used to be possible, become impossible. There will be a few cities holding on for a while to come, and some pockets of agriculture, but these will be rare, geographically circumscribed, and limited exceptions.

Again, what I’m saying is, the unknown changes brought about by Collapse will be regionalized, as will the response and adaptation to them, as opposed to any uniform “results.”

Very true, but there are also some basic things we do know, and there are the implications of those things. We do know that collapse will mean less complexity, and since we know that we can only farm now because of our current level of complexity, we know that a reduced level of complexity will mean the end of farming.

While land has primarily been exhausted for agriculture, how do you see that translating as far as pasturalism goes?

If there are low enough population densities then even a degree of sedentary pastoralism may be viable. Certainly cleared land that is too poor for farming is often viable as pasture.

In the same way that there were some primitive cultures that lived primarily on meat, could that not tranlate for post collapse pastoralists with a bit of foraging on the side?