Guns, Germs and Steel is an excellent book. Collapse is better.
When "Guns, Germs and Steel" first came out, I was fortunate to take part in a graduate seminar that was built around it. Along with reading a chapter each week, we also read a number of additional readings, some of which were technical papers.
Careful reading of "Guns, Germs and Steel" reveals that, contrary to some opinion, it is not a work of "environmental determinism". Much of history is microhistory. There is also quite a lot of history studying how quirks and idiosincrasies of important individuals shaped history. There is also quite a lot or work on the social determination of a society's success: its history, political organization, social structure etc. What Jared Diamond attempted with "Guns, Germs and Steel" was to fill the glaring gap - the dearth of studies on effects of geography, climate and environment on the success of societies. He most definitely did not dismiss the social half of the story. He just tried to provide the other half - how geography, climate, climatic change, soil quality, orientation of continents, orientation of mountain ranges, availability of domesticable plants and animals, etc. influenced the success of societies in different parts of the world. The emphasis on environmental factors was neccessary for a book designed to fill this gaping chasm, yet I can see how a quick glance at the book (not even to mention here the biased or dishonest reading) may mislead one into assuming that the book is a work of environmental determinism.
But there are always people who voice their opinions about a book without reading it. Worse, there are many people who invent arguments out of thin air and dismiss the real experts a priori.
I am afraid that "Collapse" will suffer the same way, due to people's tendency to skim books shallowly. It is not environmental determinism. It is not a tree-hugging manifesto. It is not written from an Eurocentric perspective. It does not fall for the "noble savage" myth. Yet I have already seen all four of these accusations leveled at Diamond on blogs and in online book reviews (sorry, I will not provide links to those places). So, Jared Diamond, smartly, makes sure that (at least a rare careful) reader of "Collapse" is reminded every few pages that the analysis is not simplistic. The Introductory chapter is a clearly laid out statment of methodology - well worth reading on its own. The book is a maddeningly unbiased, multifactorial analysis of the complexity of factors that determine which societies collapse and which succeed. No-no-no, it is not unreadable academic dense prose. Au contraire - it is laid out clearly and written in an easy flowing style, so do not be afraid of its bulk. It will draw you in and you WILL read it through.
While "Guns, Germs and Steel" studies how successful societies get started, "Collapse" tries to figure out how societies manage to persit for long periods of time and why they collapse once they do. He compares a number of past and current societies on a whole host of factors. Some of the factors are purely environmental (e.g., geography, climate, climatic changes, soil quality, geographic isolation, rate of natural soil renewal, amount of rainfall, temperature, spontaneous environmental changes etc.). Some of the factors are purely social (e.g., land of origin before immigration, history, social identity, social organization, political system, religion, customs and habits, scientific and techonological prowess, quirks and idiosyncrasies of powerful people, trade relations with friendly neighbors, war relations with unfriendly neighbors, etc.). Other factors are a combination of both (e.g., effects of environment on human activity, effects of human activity on the environment, responses of the society to environmental change, etc.). He details the histories of a number of societies, exploring the relative importance of each of these factors and how those factors feed back (and sometimes feed forward) on each other, i.e., how they are NOT additive but interactive. Not once he singles out just one factor to claim that THIS is what doomed Society X.
Thus, in a biological parlance, his study is INTEGRATIVE. It is also COMPARATIVE. First of all, he examines a number of cases, some of which are past collapsed societies, others past successful societies, yet others are modern societies with various problems. Even better, he manages to do some pair-wise comparisons as well. For instance, he compares Greenland Norse to Greenland Inuit (to Iceland to Shetland); Haiti to Dominican Repubic; Rwanda to Burundi; Mangareva to Pitcairn Island; Tikopia to Tonga; New Guinea to Japan, which are examples of pairs of societies that have several factors equal or very similar (i.e., they are, in a sense, "controlled for"), yet often have different outcomes, thus provide good case studies for elucidating key differences that led to different outcomes.
Diamond never says it explicitly, but one of the take-home messages that I got was that it is only very successful societies that suffer spectacular collapses. Societies that are not very good at solving problems and providing for its members never manage to get rooted in the first place - they are mere blips in history. On the other hand, soceties with social structure, political organization and rates of technological innovation that make them very successful, collapse due to the very strengths that allowed them to survive for so long. The solution to the problems that environment poses to them led to relatively easy and good life, which in turn led to population growth. The large population may still have been OK during good or average years, but if there was a series of bad years (e.g., very cold or dry), they may have exceeded the carrying capacity of the environment (Maya and Anasazi being perhaps the best examples of this). The same types of "bad years" that they easily survived when they were smaller in the past spelled doom once they got big - technological gizmos nothwithstanding. Also, the very type of social and political organization that made the society succesful in the beginning may spell doom once the population is too large. The initial strengths become liabilities once the situation changes. And there is no one-size-fits-all recipe for success.
Diamond powerfully argues that a top-down political organization works in some situations (e.g., large societies, like Japan and China) and a bottom-up approach in others (e.g., New Guinea, Tikopia, i.e., very small societies, but also, as exception to the rule, a very large one like Australia and, presumably European countries and the USA), and a mix of approaches in others (e.g., Dominican Republic which is a mid-size society). I prefer the bottom-up approach myself, but he persuaded me, against my will, that it would not work everywhere.
The very word "collapse" has a negative conotation, yet Diamond writes with admiration about many of the extinct societies. Sure, they collapsed, but that was after very long times of very successful management of problems their environments posed. Neither it neccessarily means that every single member of that society died. There are still natives on Easter Island, yet the hight of glory passed along time ago.
The biggest part of the book - three whole chapters - describes the collapse of the Greenland Norse. Diamond rightfully stresses several times that Greenland Norse were the most successful society to ever inhabit Greenland. Several attempts by Native Americans to settle there failed. Inuit are still there, but they have been there for only about 200 years and have suffered many small local collapses where whole villages starved to death. The current "modern" inhabitants of Greenland survive only due to subsidies from Denmark. Only the Norse managed to survive and thrive in the inhospitable climate of Greenland for 500 years. Remember, we have not survived that long in the mild and fertile USA yet. Thus, why the Norse failed to survive is a big puzzle and a fascinating story. And no, it is not because they did not eat fish. Diamond evaluates the evidence and agrees that some Norse may have eaten some fish. But he also argues that even if the Norse adopted good fishing methods and ate as much fish as the Inuit, that would not have been enough. Perhaps they would have survived another year or two, but they were still going to die out. There were just too many factors working against them (and not against the Inuit), the non-eating of fish was just one of many factors, and not even a major one.
The latter chapters in "Collapse", those about modern societies, are emotional rollercoasters. Page after page of gloomy statistics, hair-raising examples of stupid and greedy environmental damage, ....and just when you are about to swear off the human race (my thoughts: "China is going to kill us all in 10 years" and "I'll never even go to visit Australia"), Diamond changes the tone 180 degrees and starts listing all the positive developments, especially the most recent ones, and ends on a cautiously optimistic note (and you take a deep breath of relief). He describes the future as a horse-race between the forces of environmental degradation and forces of reason that fight against it, and he is hopeful that the latter horse will win.
The last three chapters - about the Big Picture and take-home messages - are probably the best, but it is difficult to read them without reading the rest of the book before that. First, Diamond keeps refering to information from earlier chapters in the book. Second, you may not believe him if you have not absorbed the prior information. He lists a dozen major problems that we need to save within our generation's lifetimes if we want to leave our children a livable planet. All dozen feed on each other and all dozen, not eleven, HAVE to be solved. He has no qualms about naming the individuals and companies who are "bad" stewards of the environment, but he is most definitely going to enrage many tree-huggers with a whole long chapter on big well-known megacompanies that nobody likes yet Diamond describes how they do excellent job in re-shaping whole industries towards a more sustainable practice and even positive action towards rebuilding the damaged environment.
Finally, Diamond directly addresses and counters the most frequent responses he expects his book will elicit and, in the last chapter, gives advice how we, as individual citizens of our communities and the world, can make a difference (including what courses of action do NOT work).
While it is best to read both of Diamond's books as a two-part study, they can be read independently. They ask somewhat different questions and apply different methodology to come up with the answers. As much as I think that "Guns, Germs and Steel" is an excellent book and am glad to see that it has been on the NYTimes besteseller list continuously for years, I still believe that "Collapse" is a better book and a more important book. I hope it replaces something (some Hannity drivel or Coulter gibberish) on the NYTimes bestseller list for many years to come. I hope that people who buy it take the time to actually read it. I know it's thick, but it is worth the time and effort. It may change the way you think. It may even change your life.