Book Review. The Dark Forest (2008)

This book is the second book in the trilogy titled "Remembrance of Earth's Past" by Liu Cixin. I had reviewed the first book in the series, "The Three Body Problem", earlier. 

I had listened to this book as audiobook. The reader of the audiobook, the voice-actor, was very competent, and also remarkably good in speaking in accents. This made the already engaging story more captivating.

As in the first book, this book also introduces new concepts. The Dark Forest theory is one. The theory posits that the universe is like a dark forest, where everybody is out there to hunt anybody. Due to chain-of-suspicion, a civilization can never be certain of an alien civilization's true intentions.
The extreme distances between stars creates an insurmountable "chain of suspicion", where any two civilizations cannot communicate well enough to dissipate mistrust, making conflict inevitable. Leaving a primitive civilization alone is not an option due to the exponential progress of technological change, and a civilization you have detected might easily surpass your own technological level in a few centuries and become a threat. If you have detected a civilization, then you have also confirmed that said civilization will eventually be able to detect you. Therefore, it is in every civilization's best interest to preemptively strike and destroy any developing civilization before it can become a threat, but without revealing their own location to the wider universe, thus explaining the Fermi paradox.
This is very much the coordinating attack problem in distributed systems. Two generals will attack a city. Only if they attack together, they will have victory. If only one party attacks, and the other party does not join, the attacker will be defeated/destroyed. The two generals try to coordinate this attack by talking to each other via messengers, which may be captured by the city. When the channel is arbitrarily lossy, two parties cannot solve the consensus problem in a deterministic manner for all cases. Chain-of-suspicion arises.
  1. When you get my message, you know my vote, but I don't know you know. So you send me a message back to coordinate the attack, because you suspect that if I don't know you know, I won't join you in the attack.  
  2. When I get your message, I know that you know my vote. But I cannot commit on this, because you don't know that "I know that you know" (because the message may have been lost). So I suspect that because of this lack of information, you won't join me in the attack. So I send you a message back. 
  3. When you get my message, you know that I know that you know my vote. But you realize that I don't know that "you know that I know that you know my vote". Then you suspect that under this condition, I may not join you in the attack and you will get destroyed by the enemy attacking alone .So you send me a message back.
  4. When I get your message, I know that you know that I know that you know my vote. But I realize that you don't know this: "I know that you know that I know that you know my vote", and I suspect you won't join me in the attack due to this lack of information. So I need to send you a message. 
(And this is when the generals were not Byzantine. With Byzantine participants, things get even more hairy.) This chain of suspicion makes consensus/cooperation impossible. However, it should be noted that these are theoretical impossibilities and has restrictions: such as using deterministic protocol. The reality is not this bleak. In practice we solve the coordinated attack problem every time we establish a TCP connection. I had written something on this before.

This book picks up from where the first book has left. The setup is that Trisolarans can observe, through sophons, any visual and audio activity in the world, and even anything written to any electronic storage. The only thing they cannot observe is what goes on in human's brains. (What an interesting setup... I suspect the computer security researchers would appreciate this setup even more.) This led the earth to select wallfacers, the keepers of plans known only to themselves, who are granted full access to the resources of the UN. Another interesting twist was that the Trisolaran society does not have any distinction between communication and thinking, as their thinking is opaque. So they were initially unfamiliar with lying and deception. I thought the book could have developed on this more but this didn't develop as much I thought it could be.

Other interesting concepts include the droplet from Trisolarans, the hibernation to move in time toward doomsday battle, and the mental seal machine.

It was a long book but very interesting and very engaging. I really like Liu Cixin's writing. Of course the protagonists and most of the main characters are Chinese, because Liu is Chinese and is writing for a Chinese audience primarily. After all we have seen many many many AngloSaxon/Western heroes in the literature and media, and of course Chinese protagonists have every right to save the world themselves. I welcome that. But unfortunately I started seeing Chinese government party-lines seeping into the book's narrative. The first book had stroke a balanced tone with respect to this (especially with its discussion/coverage of the cultural revolution), but in this book, especially in the second part, there are undertones of propaganda. The book discusses why it is important to be cold-blooded pragmatists and how totalitarianism is very efficient to deal with crisis situations, etc. I just started the third book in the series, and I find that there is even more propaganda in that book. But, of course, that won't stop me from reading and enjoying the book.

I like these book series because it sets the perspective to a wide perspective: the infinite space versus the marble sized earth. It is a shame we are not thinking more about space, and establishing space-fairing civilizations. When I was growing up in Turkey, I had a very exaggerated view of Turkey. For me it was the entire world. I thought its institutions and its politicians were important and powerful, and the power struggles there were consequential. When I traveled to US for graduate studies, I very quickly realized how limited that worldview that was. Similarly, earth-centric thinking is very limited, but we are very slow to realize this. 

In the book, all the projects are all planned ahead for the four centuries to come. But you can appreciate that although our time is just but a little sliver of time in that timeline, the actions we take are important enablers and first steps for the centuries down the line. So yes, I really enjoyed the big perspective thinking this leads to, and I think we should try to expose ourselves to this type of thinking to break free of our very limited perspective in terms of time and space. 


Popular posts from this blog

Foundational distributed systems papers

Your attitude determines your success

Progress beats perfect

Cores that don't count

Silent data corruptions at scale

Learning about distributed systems: where to start?

Read papers, Not too much, Mostly foundational ones

Sundial: Fault-tolerant Clock Synchronization for Datacenters


Using Lightweight Formal Methods to Validate a Key-Value Storage Node in Amazon S3 (SOSP21)