This quote, which opens the main sequence of every single The Wheel of Time book, can, with some alterations be applied to reading, memory, and the impressions formed (and altered) from the commingling of the above. I am not by nature someone who trusts wholeheartedly my first impressions; too often they change with time and further reflection. I have found this to be the case with this now-thirteen-volume epic fantasy series. When I reviewed the twelfth volume (and the first where Brandon Sanderson wrote most of the material in place of the deceased Robert Jordan), The Gathering Storm, in October 2009, I perhaps was a bit too forgiving of that book's shortcomings because I reviewed the book after not having read most of the other volumes since 2000. Certainly my memory did not jibe too well with my experiences re-reading the first eleven volumes this past spring and writing commentaries on my impressions. In short, it was a slog re-reading this series. Not merely because of the myriad subplots nor because there were repetitive and yet shallow social commentaries, but also due to the creaky, non-graceful prose and uneven characterizations that often left me feeling cold. Despite the change in authors and the plot developments that one might expect in the penultimate volume of such a ponderous multi-volume series, Towers of Midnight, after some reflection, is a flawed volume in a very flawed series.

Most reviews of a thirteenth volume naturally will be the thoughts of fans (or former fans), intended more for those who are going to read the volume regardless of the reviewer's reactions than any such essay being targeted toward those who are unfamiliar with the series. When I agreed to receive a review copy and to consider writing a review of this book, I did so largely to see just what my reaction would be to the latest installment in a series that has diminished in my esteem over the past decade. Interestingly, my initial reaction was mostly a sort of backhanded compliment, something along the lines of "Oh! Sanderson has eschewed having faux bondage scenes in here! And hey! I don't have to endure the repetitive thoughts on how this male or female PoV character states their bafflement at the opposite sex! Sweet!" rather than being wowed by the mechanics of the story. However, in the interim of nearly three weeks between me receiving a review copy and the writing of this review those initial positive reactions have faded while my unease at the structure of this novel increased.

Doubtless, most fans of the WoT series are just excited to discover that "stuff happens!" It is true that on a plot level, there are several important reveals that either further or conclude several plot threads, some of which had been left hanging since the earliest volumes of this series. For those that treat this series as merely an extended Wikipedia summary, doubtless the developments here (from the starting of major combat operations up north to battles in the world of dreams and one male character growing a pair and admitting his own nature to a long-expected rescue of a character left for dead eight volumes ago) make Towers of Midnight an exciting must-read for them. But for those readers such as myself who wish there would an elegance to the writing or at least a sound structure that limits herky-jerkiness to a minimum, this book perhaps might be one of the most poorly-written volumes in a series that is infamous for its sometimes-execrable prose.

In previous reviews of his solo efforts, I have noted, if usually in passing, that Sanderson's prose rarely is more than "serviceable." There is nothing inherently wrong with aiming for "invisible prose"; I just have different preferences. However, in this volume and much more so than in the previous, Sanderson's prose inclinations clash noticeably with those scenes originally written by the deceased Jordan. Time after time, there would be several pages or chapters full of prose that is short, staccato, and focused more on getting from Point A to Point B as quickly as possible when suddenly the tone would shift (often, this would be when a character moves to a different locale) and then there would be more ornate descriptive prose. After a while, it became irritating, often because I would be jarred out of my accustomed reading rhythm just so I could process these narrative shifts before the flow would turn back to the staccato rhythms of before.

Although I am far from a superfan (or WoThead, as some call those fans who obsess over the minute details of the series), it was irritating to see continuity errors. From how some of the main characters were portrayed (particularly in their interactions with other main characters) to a few of the small details I recalled from re-reading the series a half-year ago, those mistakes in terminology, in characterization, and even in long-established plot lines, these errors contributed to the sense that the writing was uneven at best and poorly-constructed at worst. These faults are not necessarily all Sanderson's; Jordan's narrative structure had become unwieldy several volumes before, making it near-impossible for anyone to keep things 100% true to each and every one of those minute developments in plot and character. However, it is telling when they are numerous enough that they grate on someone who doesn't devote much time at all to thinking about this bloated series.

Furthermore, the character developments were often too abrupt. At the end of the last volume, the reader sees the Dragon Reborn's epiphany on Dragonmount. But when Rand appears first, he has morphed into a sort of messianic figure; there is a sense of falseness that rings in that scene and subsequent ones because it is so jarring and it goes against the grain of the previous few volumes. To a lesser degree, this is seen in Egwene and Perrin's scenes. Despite writing a nearly 900 page book, Sanderson has written a volume that lacks adequate character transitions. This failure lessens the power of the transformations that do occur in this volume, as they feel flat and less emotional than they perhaps should have been, due to the lack of development that sets up these drastic changes.

This sloppiness and sketchiness even appears in the internal chronology of scenes. Without much in the way of explanation, several subplots in Towers of Midnight are a month or more behind others that were treated in The Gathering Storm. In one memorable moment, the character Tam al'Thor, the adoptive father of Rand, appears in two distinct places (and times) within a span of a handful of chapters, with virtually nothing in the way of referring back to events in the previous volume that set up those situations. It's the little things like this that dampened my initial enjoyment of this book. Almost every subplot contains these little problems. Individually, they would not effect my reaction to the story, but cumulatively they downgrade Towers of Midnight from an enjoyable addition to the WoT story to being one of the most problematic books in the entire series. Recommended only for those who are committed to finishing this series to the (perhaps) bitter end.

This was originally posted on The OF Blog, which tends to have a vastly different audience than those who visit RAFO regularly, so keep that in mind, please