This book is a delightful feast for readers interested in languages, their history and constant evolvement. I first noticed it as a recently published audiobook on audible.com. Not only languages but dead languages -- of course, I grabbed it. When I finally started listening to it, the preface with general musings about the reason and purpose of the book worked pretty well on audio, but when it got to the gritty language stuff, and I sensed the author could be actually quoting Greek words and even sentences spelled in Greek, I just had to grab the Kindle ebook as well. So I switched to the ebook and made some 87 (!) highlights along the way.
So did I find the book interesting? You bet. Honestly, I don't know how this book, with all its delightful multi-language quotes from ancient texts, could work on audio without losing much of its content and enjoyment. I can imagine relistening to it on audio though, or listening to it while having the text in front of your eyes.
Chapter by chapter, "How Dead Languages Work" takes you on a journey through Greek, Latin, Old English and Germanic languages, Sanskrit, Old Irish and the Celtic languages (Welsh, among others) and Hebrew. As you travel through this amazing linguistic universe, you learn a lot about how languages work, how they evolve, how each language has its own peculiarities and its own way of expressing meaning, and how some of this meaning is inevitably lost or transformed in the process of translation to another language.
The tone of the book is neither of an academic article impenetrable for the general public nor of your average dumbed-down oversimplified popular science bestseller. I would call it academic-lite -- on the one hand, serious and rigorous, with lots of footnotes about sources and intricacies, and on the other never boring, filled with dry humor and easily readable for a non-specialist interested in the topic. Even the "dreaded footnotes" are often a delight to read.
I will not torture you here with my 87 highlights -- anyway, you wouldn't stay to suffer this abuse. But I can't resist attaching a quote or few for each chapter.
Preface:
It’s all very well to say that Latin poetry can do different things from English poetry because of the more flexible word order, but if you don’t know Latin, then it’s hard to get much of a feeling for what that means—unless someone takes you through some actual lines of Horace. Accordingly, the book works in turn through six ancient languages or language groups that are especially important to the West, and which illustrate the wide range of personalities that different languages can have. Throughout, the fundamental aim is to show readers just how much they miss when they read the great works of ancient literature in translation—no matter how good the translation—rather than in the original.
The chapter on Greek provides interesting insights into the peculiarities of English spelling, among other things:
Why spell with ph, ch, and rh sounds that could be adequately represented with f, k (or c), and r? The Spanish, after all, manage just fine with cronología, retórica, and física. Also, why use an h, anyway? It makes sense in a word like uphill, where there really is a p followed by an h, but not in photo, where there’s only a single f sound at the beginning. Again, the explanation lies in the change of pronunciation over time. Leaving aside rh, which is a different matter,5 take the digraphs (two-letter combinations) ph, ch, and th. When used for words of Greek origin, these represent the single letters phi (ϕ), chi (χ), and theta (θ). In Modern Greek, these are pronounced respectively f, ch-as-in-Bach, and th-as-in-theta—what linguists call fricatives, in which the airflow through the mouth isn’t completely blocked off but merely restricted. Not so, however, in Ancient Greek, when they were aspirated stops—that is, a p, k, or t (all examples of stops, that is, sounds where the airflow is completely blocked) each followed by an h sound (that is, aspiration). In other words, the th in a word like Ancient Greek anthos (“flower”) would not have been pronounced as it is in our word anthology but as in ant-hill. In late antiquity, the earlier pronunciation of these sounds as p+h, k+h, t+h would eventually weaken to the fricative pronunciation they have today—but not before Latin, which itself borrowed a lot of words from Greek, had standardized the practice of spelling them ph, ch, and th like the aspirated stops they were. But even when the pronunciation changed, the spelling with the h remained.
Most readers of this book will probably be sympathetic to the idea that poetry, in particular, loses a great deal in translation. But what about prose? It’s a more pedestrian form of language—literally, in fact, in Greek, where the adjective pezos can mean either “on foot” or “in prose”
Latin:
in purely written documents, there was potentially scope for confusion, and the identical spelling of malum “bad” and malum “apple” is what led the fruit that Eve gave to Adam to be identified in the Middle Ages as the apple. (It had been left unspecified in Hebrew.)
Old English and the Germanic languages:
the whole process whereby a word gradually goes from being a full word with its own semantic force (“I have a car”), to an auxiliary word used to mark a grammatical feature, like tense (“I have walked”), to an unstressed unit that no longer counts as a full word, but, as a so-called enclitic, leans on another (“I’ve walked”), to a ending that’s become fused with its host word (“aimerai”) is what linguists call grammaticalization.
Essentially, a past tense like I walked is the end result of the weakening of a structure comparable to I walk-did. That a fuller verb form ultimately underlies the -ed suffix can be seen most clearly in the oldest Germanic language to be attested in any sizable corpus, Gothic. The weak verb hausjan “to hear” has the following past-tense conjugation: While the exact origin of this paradigm is beyond the scope of this book, the plural endings, with their d-vowel-d structure, are excellent evidence that the past tense of these Germanic verbs was originally an early form of did
The Germanic languages developed new words for “dawn”—dawn itself is related to day—but the old word survived in a somewhat shifted semantic domain. Dawn breaks pretty regularly in one particular part of the sky, and the Germanic languages preserved the derivative *austa- as the designation of the east. (Germanic au most often becomes ea in Modern English: this is not the most transparent of sound changes, but it does account for equations like German Baum “tree” = English beam, and Laub “foliage” = leaf.) Furthermore, when Germanic speakers were searching for the best term to designate the central holiday of Christianity, the day of the Resurrection, rather than simply borrowing the Hebrew word for Passover, pesaḥ, which passed into Greek and Latin as Pascha (whence Spanish Pascua and French Pâques), they turned instead to the old root that referred to the dawn—because, if we are to trust Bede (De temporum ratione ), the festival of a pagan goddess of the dawn was held at about the same time. And so we now celebrate Easter.
Sanskrit:
The more that was known about Sanskrit in the West, the more it became clear that any account of the origins of Greek, Latin, and the Germanic languages had to take into consideration the similarities they shared with Sanskrit, the language of such important Hindu texts as the Vedas, the Upanishads, and the Bhagavad Gita. This awareness was stated most famously by Sir William Jones, in a speech delivered before the Asiatick Society of Calcutta on February 2, 1786, a date now generally given as the birthday of Indo-European linguistics.
As it happens, l and r are similar sounds to begin with, and many of the world’s languages have only one or the other. Japanese, for instance, has only r, so when foreign words with l are borrowed into Japanese, the l’s all turn into r’s: thus, the English words salary man become Japanese sarariman. So too in Sanskrit, inherited l’s generally became r’s, leading not only to puru but also to the following equation: English light (i.e. the opposite of darkness), Latin lūc- “light”, Greek leukos “white”, but Sanskrit ruc- “to shine”.
Old Irish and Celtic languages:
Old Irish, the form of Irish spoken from roughly the seventh through ninth centuries ad, is arguably the most difficult of the Indo-European languages, almost willfully so.
We’ll limit ourselves here to a couple of sentences from the most important of them, the Táin Bó Cúailnge (“Cattle Raid of Cooley”). The Táin belongs to what is known as the Ulster Cycle, an important set of tales featuring the king of Ulster, Conchobor, and his chief warrior Cú Chulainn. In the Táin, the rival king of Connacht, Ailill, and his rather more impressive queen Medb,16 mount a raid to seize a particularly fine bull that’s kept in Ulster, but their aims are repeatedly thwarted through the actions of Cú Chulainn, who acts as a sort of Irish Achilles. What makes Cú Chulainn different from Achilles—and, with it, the Táin from the Iliad—is a much more outlandish degree of fantasy in the description of his prowess in battle. He is particularly associated with the ríastrad, variously translated as “warp-spasm” or “torque”. Here are just a few of its more picturesque effects: The first warp-spasm seized [Cú Chulainn], and made him into a monstrous thing, hideous and shapeless, unheard of. His shanks and his joints, every knuckle and angle and organ from head to foot, shook like a tree in the flood or a reed in the stream. His body made a furious twist inside his skin, so that his feet and shins and knees switched to the rear and his heels and calves switched to the front … His face and features became a red bowl: he sucked one eye so deep into his head that a wild crane couldn’t probe it onto his cheek out of the depths of his skull; the other eye fell out along his cheek … [I]f a royal apple tree with all its kingly fruit were shaken above him, scarce an apple would reach the ground but each would be spiked on a bristle of his hair as it stood up on his scalp with rage.
We next turn to Welsh, today the most widely spoken of the Celtic languages. This sometimes comes as a surprise: given the size (and boisterous pride) of the Irish diaspora, not to mention the fact that the Republic of Ireland is an independent country, Irish would at first glance seem to have had a better chance of holding its own against English. But only about 100,000 people, less than 2% of the population, live in the Gaeltacht, the part of Ireland where Irish is still spoken regularly outside the classroom, whereas about 562,000 people in Wales can speak Welsh—nearly 20% of the population.
Hebrew:
One final word in this passage worth looking at is ’ôr, the word for “light”. While there aren’t a lot of direct English derivatives of the words we come across in this chapter—we’ve left Indo-European behind, after all—there’s probably a connection between this word and ancient objects referred to in English by their Hebrew name: the Urim and Thummim. To judge from passages like Exodus 28:30 and 1 Samuel 14:41, these were, it seems, some sort of item placed in the breastpiece of the high priest that were used in the determination of judgments by the casting of lots—essentially, the Magic 8 Ball of the ancient Israelites. Whatever the exact physical make-up of these objects (and, for that matter, whatever the actual etymology of the phrase, which remains disputed), the first word, ’ûrîm, was taken to be yet another intensive plural, in this case to a word ’ûr, understood to be from the same triconsonantal root, ’‑w-r, as ’ôr, the word for “light”.11 And the second word, tummîm, was regarded as a derivative of the root t-m-m “to be complete, have integrity”. Accordingly, in the Latin Vulgate of Exodus 28:30, they’re translated as doctrīna “teaching” (viewed as a type of illumination) and vēritās “truth”, following the Greek Septuagint fairly closely, which offers dēlōsis “clarification” and alētheia “truth”. Somewhat less expected, the phrase also makes an appearance as the motto of Yale University, which both depicts the Hebrew letters (’ûrîm wə-tummîm) across the pages of an open book and the Latin translation Lūx et vēritās (“Light and truth”) below—a gloss on the Hebrew that, while an excellent guiding principle for a university, doesn’t do justice to the semantic complexities of the original phrase.
Epilogue:
Languages are embedded in the societies that speak them, and the narrative thread can be woven more tightly if the connections drawn between the sample texts include not only specific points of grammar but also shared motifs, like the dragon-slaying myth, and the broader influence one language can exert on another, with Lucretius extending what Latin can do as he proselytizes for Greek ideas, or the translators of the King James Version (KJV) repeating the phrase It came to pass in order to reflect Hebrew syntax.
The last sentence is worthy of a true language study enthusiast:
What emerges from all of these works is the sheer wealth of material dead languages give us to explore—more than one can possibly get to in a lifetime. And whether you end up chasing down any of these leads or not, I hope you’ll excuse me now: there’s some Greek I’d like to go read. Or maybe it’s time for some Latin. Or Hebrew. You get the picture.
I'm very glad I stumbled upon "How Dead Languages Work". Now Kindle keeps suggesting I should read "Learn to Read Ancient Sumerian". I can't deny I'm tempted by this daunting enterprise.
Read in 2022.