There is something very unsatisfying about The Fifth Head of Cerberus.
As far as SF goes, it leans more towards the “speculative” than “science” end of the dual abbreviation. I’ve heard praise about the lack of infodumps prior to reading, and it’s mostly correct. The lack of details was not the problem here. The too blatant deliberate obfuscation was.
I’m not sure what exactly The Fifth Head wants to do, but I'm sure it's trying to be very, very clever. Trying is the key word here. It’s trying so hard. To be deep. To be elusive. To be revolutionary. But I’m just not seeing any substance under its rather transparent style.
There are two main themes: identity and colonisation. It made a point about colonisation, but I felt it failed to make one about identity.(show spoiler)
I found the writing, particularly the construction, problematic. It is composed of three novellas, loosely connected. The style varies wildly. If these were novellas that were written separately and then gathered in this book, I would have understood. But what happened is that Wolfe wrote the titular first novella in the collection, and then was commissioned to write more so they could be published as a book. Now, varying styles would not have been bad in theory, but...something just feels jarring. Again, I simply cannot shake the feeling that the author was really trying to make them seem like different stories, and then link them in most obscure ways. I was reminded of Life of Pi, which was also divided into three segments. I didn’t mind it so much there, and I loved that the third segment was the interview script. A similar technique is applied in the third novella of this collection as well. However, while Martell keeps the story concise and restricted to a long, uninterrupted interview, Wolfe is not satisfied: he has to mix a journal, tape recordings and add mostly unnecessary snippets about life of the officer that is responsible for the review of abovementioned. Again, this might have worked if things were linked more smoothly; but the order in which we keep skipping from tapes to journal to the officer trying to secure the camp bicycle for the night’s enjoyment feels purposefully random. Probably in order to “hide the clues”. But is the answer of the puzzle worth it? Not really, in my opinion. The puzzle of colonisation is pretty straightforward. The puzzle of identity is ultimately pointless. The puzzle of the title is, frankly, trite.
This is hardly my first meeting with metafiction. I’ve enjoyed The Dictionary of Khazars well enough, as well as just about anything by Zoran Zivkovic I’ve read. None of these works offer an answer to the questions they ask; but they aren't truly asking for it, either. However, they do know how to immerse the reader into the meta-experience. The Dictionary can be read in any order you like; it presents the same story from multiple points of view, in fragments. But they are all obviously linked and the style is consistent. It’s a mosaic where every piece supplements another. As for Zivkovic, it always felt to me like his stories take place in the same world, like ours but not quite; a Twilight zone, if you will. He is even more abstract than Pavic; his prose and imagery is what creates the link, drawing the reader into his dream-like world. As a matter of fact, his stories might have as well been someone’s dreams. In comparison, The Fifth Head is a very (deliberately) messy room where we’re supposed to hunt for clues under the junk.
I suppose it all depends on you, potential reader. What is meta for you? Do you want to be asked questions to contemplate slowly, enjoying the pondering without rushing to get the results? Or do you want to dig for the answers there and now? Do you prefer slow stewing or instant gratification?