Ancient Romans couldn’t read… and what that means for language

Agricola servat nautam.

That was the first Latin sentence I learned. “The (or a) farmer saves the (or a) sailor”. The point of using this sentence straightaway (written in a textbook used across the British Isles whose author was Head of Latin at Methodist College, out of interest) was emphasised by the next one:

Agricolam servat nauta.

This means “The (or a) sailor saves the (or a) farmer”. The point is that it is the word endings which contain the grammatical information as to who acted (the subject) and who was acted upon (the object). With the exception of personal and occasionally relative pronouns (e.g. “he saves him“, “the sailor, who(m) he saves”), modern English and nearly all languages derived from Latin no longer use the word endings to provide this information – instead they use word order. Modern English, alongside most national languages derived from Latin (French, Italian, Spanish, etc), are all known as “SVO” languages – generally, the subject (“doer”) comes first, then the main verb, then the object (“done to”) – “l’agriculteur sauve le marin”.

Furthermore, the ending could also be used to indicate other information: nautae is “to/of the sailor”; agricolā is “by/from the farmer”, and so on. So “Servo agricola nautae” may mean “I save the farmer with the sailor”. As a result, fewer prepositions were employed by the Ancient Greeks and the Classical Romans.

As it happens, Latin was generally an “SOV” language (thus the most usual form was “Agricola nautam servat“), which is in fact the most common type. Hints of this remain in modern French, Spanish and Italian where, for example, pronoun objects precede the verb (“Je t’aime”, “(yo) te quiero“, “(io) ti amo“) – literally “I you love”). Unusually, German and Dutch are “V2” languages, where the verb comes second in the sentence regardless of what came first (Old and early Middle English were like this too, and Modern English contains vestiges – “So am I”, “Hardly had I arrived…”, etc). However, the main point is this: Latin generally determined subject and object (and other such grammatical information) by word ending, whereas modern languages derived from it (as well as English, which is not) do so by word order.

Why did this change?

For me, the likeliest reason is this: the Romans couldn’t read.

Well, some of them could read – obviously. However, they could not read other than aloud. If you were transported through the Time Tunnel to Ancient Rome, you may glance over and see a Roman reading a tablet, and you would immediately notice him mouthing the words as he did so. In Classical times, speed reading was unknown. Reading itself, after all, was a comparatively new “technology”. Just as it took 500 years to move from the word on the page to the word on the screen, it took many centuries to move from reading an alphabet at all, to speed-reading it in the head without having to read it aloud (or at least mouth it).

Speed-reading in the head is clearly a much different thing from reading something aloud. Mouthing the words as they went, the Romans (and the Ancient Greeks and others, for that matter) had time to consider the grammatical information contained in word endings as they went along. They needed fewer words in total, because the endings showed how they went together without to the need to worry about word order or too many prepositions.

Once speed-reading became the norm – where small words can be skipped and little attention paid to endings – it became more useful to add more prepositions and determine grammatical relationships by word order.

We can tell from the writings between the Roman Republic and the Fall of Rome that this process began in the first few centuries after the birth of Christ; so it is possible that speed-reading began in that period, perhaps from around 200 AD. However, we cannot be sure of this, as we also know that the letter “-m” after a vowel at the end of a word came not to be pronounced in Latin from about that time, restricting grammatical distinctions in speech regardless. In fact, it is even possible that changes in pronunciation actually enabled speed reading, as it became possible to pay less attention to word endings and easy to use the extra prepositions as “signposts” while reading quickly.

A not dissimilar process may have happened in Germanic languages (though none was written down at all until 350 AD). Modern German may still express grammatical relationships by endings to some degree (so “Der Bauer rettet den Seemann” meansThe farmer saves the sailor“, but “Den Bauer rettet der Seemann” means “The sailor saves the farmer”, at least technically). Other Germanic languages (and German, in modern practice) now depend almost entirely on word order, just like those derived from Latin. The process may in fact have started slightly sooner, as Germanic languages are marked by having shifted the stress on words to the first syllable of the word (whereas Latin-derived languages it tends to be on the penultimate), meaning endings were pronounced less distinctly in Germanic even 1700 years ago. This has left Germanic languages not reliant on endings even for verbs, whereas most Latin-derived languages still are (with the notable exception of French, precisely because it was heavily influenced by Germanic languages in its early development separate from Latin). Thus, for example, Latin-derived languages except French tend to allow the dropping of pronoun subject whereas Germanic languages and French do not – hence “salvo” in Italian and Spanish (and “servo” in Latin), suffice on their own to mean “I save”, whereas the pronoun is required in German (“ich rette“) and French (“je sauve“) as in English.

However, the process of moving away from reliance on word endings generally did not occur in Slavic languages, where they are still an essential part of each language (which makes them very difficult to learn for most Western Europeans). It is unclear when Slavic speakers began speed-reading or even whether they do – it is not as necessary, after all, because there are fewer prepositions and word endings remain crucial to understanding.

Oddly, this Slavic linguistic conservatism led to the inclusion of an object ending in the well known constructed language Esperanto (thus the above would be “la konstruisto konservas la mariston” versus “la konstruiston konservas la maristo“), whose creator Ludwik Zamenhof grew up around Eastern Europe surrounded by Slavic and Baltic tongues – it would be unthinkable for a Western European to have included such a distinction. (It is perhaps noteworthy that the nasal suffix, either -m or -n, is common as a marker of the object across almost all Indo-European languages and would have existed even in Proto-Indo-European, the common predecessor language of Latin, Proto-Germanic or Old Church Slavonic, spoken many millennia ago – thus agricolam, den Bauer, and even konstruiston; note also even English him, whom, them).

This is yet another example of how linguistic development gives us a fascinating insight, in and of itself, into historical  and even technological change – and perhaps into how similar, ultimately, we all are.


3 thoughts on “Ancient Romans couldn’t read… and what that means for language

  1. Eileen Cairnduff says:

    Really enjoyed reading this blog today. Languages, and their development, are very interesting. I have always wondered why French kept the subject pronouns among the Romance languages.

  2. amghobsmacht says:

    Brilliant Ian, you’ve given a background to a problem I’ve tried to overcome and explain to others (Slavonic grammar).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: