Not recommended

I have been reading Birgitta Hoffmann’s The Roman Invasion of Britain. Archaeology Versus History (2013), which was reissued as an eBook in 2019. It isn’t an easy read. For one thing, English doesn’t appear to be the author’s first language; and for another, the Pen & Sword Books editor has turned a blind eye to grammar and punctuation, so that some sentences must be re-read before the meaning becomes apparent.

The English is often stilted (e.g. translations “for” ancient sources, costs “born” by individuals), obvious mistakes have not been corrected (e.g. a reference to “the capture of Claudius” is clearly meant to be Caratacus; a reference to action by “the Roman Empire” clearly means the Roman army; Suetonius is credited with biographies of “the Emperors from Caesar to Trajan”), and infelicities abound (most noticeably chopping and changing between past and present tenses). But, copy-editing apart, is Dr Hoffmann’s argument sound?

When is a solution a ‘valid’ one?

Dr Hoffmann’s purpose seems to be to cast doubt on everything recorded by the ancient literary sources. “Most of the so-called facts of the history of Roman Britain cannot be independently confirmed and are thus open to source criticism as well as general doubts as to their veracity.” Archaeology doesn’t get let off lightly either. “The best any researcher can hope for in reconstructing the past is a ‘valid’ interpretation.” But what constitutes validity?

Here is an example. In the context of the Claudian invasion, Dr Hoffmann refers to “a number of troops that can swim in armour through rivers” and complains that everyone assumes that they were Batavians although “Cassius Dio does not mention them: the troops that do the swimming are called Keltoi (Celts), a cultural attribution that is rather doubtful for the Germanic Batavians, but also a name that is not used in the official naming of these units”.

In the process, she claims that “E. Cary’s Loeb translation actually calls them Germans (which is not in the original Latin)” — except that, firstly, Cassius Dio was writing in Greek, and secondly, he customarily uses the word Κελτοί (Keltoi) to indicate the peoples around and beyond the Rhine; in other words, the Germans. (The passage is Dio’s Roman History 60.20.2, if you wish to look it up, and I would recommend Mark Hassall’s 1970 paper on “Batavians and the Roman Conquest of Britain” for a much clearer discussion of this episode: Britannia Vol. 1, pp. 131-6.)

Having recommended that we strive for validity (defined as “any statement that takes account of all pieces of evidence as they are known at the time”), she suggests that the swimming troops “may actually have been British troops fighting on the Roman side. There is little to refute the argument” — what about the fact that Dio calls them Germans, and emphatically calls the Britons Βρεττανοί (Brettanoi) — “but equally little to support it. Our sources are just not good enough to come to a decision, but it remains a valid solution”. No, I’m afraid it simply doesn’t.

An ancient Baedeker?

A recurring theme is Dr Hoffmann’s criticism of geographical vagueness in the ancient sources (a vagueness that’s surely quite understandable, in the absence of major towns or landmarks for a Roman audience to latch on to). Again in the context of the Claudian invasion, she is affronted that “of all these achievements [attributed to Vespasian by the biographer Suetonius], only the Isle of Wight can be identified on a map. (...) Other desirable topographical information is lacking or abbreviated.” Then, regarding Tacitus: “he certainly has the tendency of focusing on people rather than events or places, and his geographical details, in keeping with historical traditions in Rome, are sketchy, as they were deemed of little interest to the readership.”

And yet, when Pliny the Elder writes about the vicinitas silvae Calidoniae (“neighbourhood of the Caledonian forest”), Dr Hoffmann feels that this is a sufficiently detailed geographical marker to indicate the edge of the Highland massif, since (she claims) the word silva (Latin for a wood or forest; see Oxford Latin Dictionary entry, pictured here) “can also mean mountain range.” No, I’m afraid it simply can’t. (The Pliny reference is Natural History 4.16/102.)

References?

Finally, regarding this book’s usefulness as a reference work, Dr Hoffmann’s approach to her sources can only be described as casual. She uses phrases like “several recent suggestions ...” or “there have been numerous proposals”, with no further clues.

The bibliography is a little random and not entirely error-free (e.g. she has listed several “Hodgson 2009”s without differentiation). And in a book intended for Anglophone readers (and presumably archaeologists, at that), it is frankly absurd to send them to an obscure German edition of Tacitus (Wilhelm Bötticher’s nineteenth-century edition, revised in 1985 by Andreas Schäfer), particulary when the point at issue (the emendation of Annals 12.31 from castris Antonam to the universally accepted cis Trisantonam) was the work of an English scholar. (I say obscure, because — although it is listed as the sole German “scholarly translation” in a German-language reference work, Der Neue Pauly’s 2007 survey of ancient authors — it did not merit inclusion in Herbert Benario’s “Recent Work on Tacitus: 1984-1993”.)

It’s difficult to know who the target readership is. Someone who knows what “Arretine Samian” or an “Alexandrinian diobol” (sic) is. But equally someone who won’t mind that Hoffmann has confused Hyginus (author of Fortifying a Roman Camp) with the surveyor of the same name. Or who won’t notice that, when she insists that Dio’s use of τριχῆ (trichê, “triply”) is “unusual (...) a term that comes from epical and poetical Greek and is certainly not a technical term”, she is wrong (the term is happily used by Herodotus and Xenophon, even Arrian — what better endorsement could there be?). Finally, it is peculiar to see Sir Ronald Syme’s classic work of 1958, the two-volume Tacitus, referenced as “Syme 1963”.

Postscript

Besides the generally poor editing, my enthusiasm for Dr Hoffmann’s book was further dampened by the fact that an unknown quantity of text was evidently missing from Chapter 1 (pictured here, along with an example of the shoddy copyediting).

When I contacted the publisher, I was offered a replacement copy at the knockdown price of £8.00. Needless to say, I did not dignify their impertinence with a response.