Tau Zero was published in 1970 - I must have read it shortly afterwards - and tells the story of an interstellar starship, a Bussard ramjet, which suffers damage on its first interstellar voyage. As a consequence, the starship cannot slow down and the story explores the consequences as the speed nudges ever-closer to the speed of light and the crew's subjective time (the proper time, τ) tends to zero (hence the title).
So Tau Zero was lying around Alex's flat where we're currently staying, a library book waiting to be returned. As I re-read it, the most discordant thing was Anderson's shaky grasp of the science.
As the ship's speed approaches that of c, Anderson believes that the star ship gets ever more massive from the universe's point of view (correct) and that as a consequence it can 'punch through stars' with impunity (no way!). In reality, the starship’s impact with any kind of gas cloud would be the ultimate Large Hadron Collider run: the end-result a massive quark-gluon plasma trail.
Anderson’s plot requires an oscillating universe model with the universe collapsing back to a high-density state. In the story, this is not the collapse of space-time (and therefore something which happens 'everywhere’) but a collapse within a relatively unaffected space-time at a particular point. I think this misconception is necessary to the plot but it isn't right. As an additional error, he seems to think that the collapsed state is cold and arid rather than extremely hot and dense.
Does any of this matter? He gets lots of other things more-or-less right: the ship’s relativistic view of the universe (the 'starbow'); time-dilation and length contraction - although it's not directly visible as he suggests, and the staggering scale of the universe in both space and time. However, in science-fiction you are meant to get the science right so I think that's a down-mark.
The next thing I was interested in was his characterisation and descriptive writing. Here I was quite impressed (rather than, say, fantastically impressed). His main characters, Charles Reymont the constable or enforcer, and Ingrid Lindgren the empathic first officer are clearly drawn, recognisable types (ESTJ and ENFJ respectively). Chapter 1 introduces them amongst the sculptures of the Millesgården in Stockholm and the scene is convincingly and movingly drawn. Later on we meet other stock types, the bluff engineer, the semi-Asperger astrophysicist: the plot advancing through dramatic scenes which exhibit their psychological type dynamics.
As a story, Tau Zero is clearly plot- rather than character-driven. The characters are cleanly and plausibly moved around to meet crisis after crisis. Only occasionally does this create fracture lines in the narrative, where the reader wishes that some of the people described might display reactions more complex, ambivalent and nuanced - more real - than those merely required for dramatic tension and its consequent resolution. But hey, this is science-fiction we're talking about.
A novel where all concerned are confined for almost the entire duration of the novel to a spaceship is hard to keep interesting. Anderson's book is readable if not totally compelling and is chiefly notable - like most hard-SF - for the sense of awe its concepts induce. On my second reading, thirty five years after my first, that sense of awe is still there.
Poul Anderson died in 2001.