We all felt it, June 15 of this year: a 5.2 temblor 40 miles off the coast Coronado. That first shake of the building (if you were indoors); the recognition, "It's a quake"; then the peak of the seismic wave jolting the walls and the table and everything on the table. All of it in five seconds.

What I recall of that long instant is how time distended under stress. For example, the fourth of those five seconds, when the quake got much stronger. Suddenly, that unwieldy fact got me up and headed for the door; when I stepped outside, the rattling stopped. And yet how many of us think back and say, we're certain we had the presence of mind to handle whatever would have happened? A total prevarication. In the moment, you don't know how long the possibility of the quake is, which is really the possibility of your death – a fear no different, I assume, from the fear that rises in battle.

Such recent shaking in the Southland is serendipitous for David L. Ulin, and sales of "The Myth of Solid Ground." In fact, I'm sure he'd understand the elongation of time many San Diegans experienced. That sensibility is central to this dramatic work of journalism and reflection, of science and seance: Ulin wants to discover for himself what seismology and other forms of quake prediction have to say about the phenomenon, and why they (still) add up to nothing definite.

His fascination began in 1980 with his first temblor. Having returned to San Francisco after an Easter dinner in Marin County, the 18-year-old Ulin felt "a slight pitch and yaw," then "the whole room started to rock, gently." When the cups started to dance on the shelves, "time had stopped" and "the simplest things ... had been suspended." Much like our June quake, Ulin's 4.5 was "big enough to rattle nerves and windows." Something else moved through his psyche that day: a wave with one end attached to California, the other to his mortality.

It was, he writes, "a reminder of the ephemerality of existence ... as if, under the surface of this placid Sunday, there was nothing you could count on, as if ... California itself existed in a state of elaborate balance, equally solid and insubstantial, between the quotidian landscape of daily living and the explosive possibility that lay beneath."

Writing this book two decades later, Ulin discovers that he had misremembered the event. Having researched it, he learns day and place were wrong – it was really a 3.5, and it occurred the next day, while he was across the bay in Emeryville. This misremembering is indicative of the temporal dislocation that quakes bring. They become apocryphal, change via the telling, equate fact to fiction.

Predicting earthquakes is of course an inexact science, and to unpack its enigma Ulin must work both sides of the equation. He profiles seismologists and predictors, pseudo-scientists and earthquake "sensitives" – people who, like songwriter Carole King, feel the earth move under their feet hours before the rest of us do. He also includes much personal reflection, pondering his experience as a reporter (often with his young son in tow) and the unfinished nature of the evidence he combs through.

Ulin's is probably the only sane way to write about quakes. But it is surely not safe. Like abusive childhoods, temblors and their damage create a consciousness about themselves that yield to analysis. The human effect of quakes – whether to study or to experience them – is psychological.

A temblor also enthralls us on the physical plain. For example, its importance is inversely proportional to the time it takes to happen. A powerful quake releases in seconds multiple stresses in the Earth's crust that have collected for years. Quakes come in all sorts of intensities and timings: as Big Ones, 8.0 and beyond, of which the real Big One on the San Andreas remains overdue; as aftershocks that soon follow a quake; and as foreshocks, a new frame that suggests large quakes may be predicted by the power and frequency of smaller quakes before a larger one – but only if the fault system has been rightly mapped, which many haven't.

(The foreshock comprises a new worry: was our June quake merely announcing the real wall-buster coming this summer?)

The other leg of our kindergarten knowledge about temblors is that the Earth is five million years old, while Richter scale data, which measures the magnitude of released energy, is not quite 70 years old, and the revolutionary idea of plate tectonics is half that. My money is on the Earth not yielding up its geologic patterns of seismic call and response any time soon.

But that hasn't stopped the scientist and the sensitive from trying to find some model. Since scientists are seldom comfortable with prediction, sensitives abound. Among the most renowned of the California predictors is Charlotte King, who claims to "hear" the ultra-low frequencies of quakes arising from the deep in the Earth. Despite making thousands of predictions, it's said she's right 85 per cent of the time, including the Landers and the Loma Prieta quakes. Kathi Gori has accurately felt some 20 quakes "by relying on headaches that come and go a few hours before a quake."

Somewhere between predictor and amateur scientist is Jim Berklund. By using data from the moon's gravitational pull on the Earth's tides, Berklund has "found" the earthquake "window" – the six days that follow a high tide. Then there's Zhong-hao Shou, or Cloud Man, who photographs unique "finger-pointing" clouds, which, he says, indicate where the quake will soon occur: The cloud is pressurized steam that has built up in the very small pre-quake cracking of a fault and escaped. Predictors like Shou animate the antic notion of "earthquake weather," that very dry, bright, hot, still day that nudges us to note the location of the exit or the sturdiest table.

Most of these sensitives and predictors have been tracked by earthquake documentarian Linda Curtis at the United States Geological Survey office in Pasadena, who tells Ulin, essentially, surprisingly, that data is data, regardless of the source. She says that while sensitives and predictors are frequently accurate, they almost always, in their extremism, make the facts conform to their theories.

Ulin digs into her records and discovers that the anecdotal lore about quakes is as reliable as it is cryptic. It has been shown that, frequently, three hours before a jolt, birds stop singing; snakes come out of hibernation; dogs will exhibit unfocused anxiety; and near the epicenter of large temblors, spring water will pump out of the earth three times its normal rate. (This last fact was shown by a scientist who was observing fault conditions in the hills above Santa Cruz the day the Loma Prieta quake struck in 1989.)

Seismologists, a rather dour crew, are also interested in prediction, but they hate being called on for answers when they receive few grants to study fault systems. Their general conclusion is that earthquakes are scientifically unpredictable because our time frames are too recent. Some, however, posit canny theories, each of which (theory and theorist) Ulin features.

There's the elastic rebound theory, which says that because the movement of the Earth's plates causes so much strain, periodically a breaking point is reached, an earthquake occurs and the plates rebound to a state less strained. The dilatancy-diffusion theory suggests before a quake innumerable tiny cracks appear along faults, and these cracks dilate the surrounding rock. Once the rock has swelled, water diffuses into the cracks, pressing on and weakening the rock until the rock shakes under the pressure.

As quake-chaser Ulin follows the San Andreas (which runs for 650 miles, from the Salton Sea to San Francisco), he essays on chance and likelihood, he puts his ear literally to the ground, he spends many hours in the seismologists' labs (sometimes letting them talk too long) and he finds an answer in finding no answer: Neither the scientist nor sensitive has his finger on the pulse. It turns out that what we don't know and why we don't know what we don't know about quakes still forms the heart of the story.