In this final chapter, we apply the quantitative knowledge about earthquakes we have developed along the previous chapters to the subjects related to the occurrence of earthquakes, which we presented in Chapter 2, as an introduction, in a descriptive form. First, we deal with the distribution of the occurrence of earthquakes in space and time and with their magnitude. This falls under the general term of what is called seismicity, at a global or regional level. Second, we present the relation of earthquakes and their mechanism with the dynamics of the Earth's lithosphere and plate motion or seismotectonics. Third, we address briefly the problem of assessment of seismic hazard and risk and of earthquake mitigation and prediction.

The spatial distribution of earthquakes

As we saw in Chapter 2, earthquakes are distributed not on an aleatory form on the Earth surface but grouped at seismic active zones, while other zones are practically free of them. The development of modern, more reliable seismic instrumentation (Chapter 3) and methods for the determination of earthquake location, origin times, and size (Chapter 16) have led to a better understanding of seismicity, that is, the distribution of earthquakes in space, time, and magnitude. The global picture of earthquake locations shows that in great part they coincide with the plate boundaries (Section 2.7, Figs. 2.3 and 2.8). Intermediate and deep earthquakes occur at subduction plate boundaries where the oceanic material of lithospheric plates is pushed into the mantle (Section 2.7, Figs. 2.4, 2.9, and 2.10). As a general rule, at oceanic ridges there are only shallow earthquakes, whereas for island arc regions there are shallow, intermediate, and deep earthquakes (Figs. 2.3 and 2.4). Although shallow earthquakes are classified as those less than 30–60 km deep, most of them take place at depths less than 20 km, that is, in the upper rigid part of the crust that can break by brittle fracture which is called the seismogenic layer. Earthquakes at greater depths correspond to zones where this rigid and relatively cold material of the upper crust is introduced into the upper mantle by lithospheric subduction, where earthquakes may reach depths of up to 700 km (Fig. 2.10).

THE SPIRIT OF MARATHON

Mill and Runciman exaggerate: Marathon, fought on the east coast of Attica in 490 bc, was not even the decisive battle of the Persian Wars, still less of British or European history. Yet Graves’ mischievous poem is wrong too. Marathon was more than a ‘trivial skirmish’. True, if Persia had won Athens would have survived, and the returning tyrant Hippias would still have had a city to rule. The fate of Eretria (101.3n.) shows what would have happened. Temples and sacred places would have been burned, as they were to be in 480 (8.50, 53); some citizens, especially perhaps the best-looking boys and girls (cf. 32), would have been deported to Persia to make good Dareios’ threat and instructions (94.2), but by no means all – the ships would only take so many; most important of all, this would count as ‘enslavement’ (94.2) to the Persian king, with the blow to human self-respect that this meant.

Throughout much of the twentieth century, modern linguistics suffered from a written language bias (Linell 2005): the general thinking about language, as well as the descriptive concepts and categories developed for linguistic research, were tailored to language as it appears in writing. Interactional Linguistics, by contrast, grew out of an interest in spoken language and a desire to see it studied in its natural habitat: in social interaction. Interactional linguists took seriously the observation that “some of the most fundamental features of natural language are shaped in accordance with their home environment in co-present interaction, as adaptations to it, or as part of its very warp and weft” (Schegloff 1996b:54). They pursued this observation to its logical conclusion: namely, that as a consequence language must be analyzed in the home environment of co-present interaction. The idea was to apply the same empirical methods that had been so successful in revealing the structure and organization of everyday conversation, the methods of ethnomethodological Conversation Analysis, to explore the structure and organization of language as used in social interaction. From the cumulative results of such exploration, it was hoped, a new theory of language would emerge.

Naturally, in order to explore language in its home environment, Interactional Linguistics needed and continues to need an appreciation of the structure and organization of interaction. For this, it is heavily indebted to Conversation Analysis, whose practitioners over the years have laid bare the structural underpinnings of coordinated interaction in social contexts. However, as the “founder” of Conversation Analysis, Harvey Sacks, explained early on, he was not interested in ordinary conversation because he had any “large interest in language,” but rather simply because it was easy to record and because these recordings could be listened to again and again (Sacks 1984:26). Conversation analysts are primarily interested in understanding how interaction works, in uncovering the mechanisms of, for example, how turns at talk are coordinated, how actions are constructed and recognized, and how they are made to cohere in sequences of interaction. Interactional linguists, by contrast, do have a “large interest” in language, because they believe that turns, actions, and sequences are accomplished and made interpretable by the systematic use of linguistic resources; consequently, they make the linguistic forms deployed by ordinary speakers in everyday and institutional encounters the focus of their attention.