Study smarter, not harder using word frequency
My high school ex-boyfriend and I had a frequent argument. I, ever the rebellious one, almost always refused to study for our AP Latin tests the way an AP Latin student probably should. Don’t get me wrong; I certainly needed to study – I had skipped an entire year of Latin and missed 80 days of my sophomore year due to a mysterious illness. But I was honestly much more interested in ~miscellaneous~ fandom than in De Bello Gallico.
My high school ex was interested in success. I have to hand it to him; he was always willing to do whatever it took to catapult him to unknown heights. But he lost countless hours of his life in the process, tediously memorizing figures without understanding how they related to broader themes. (OK, that was my biased opinion, but I feel like if you were 16, you’d be pissed too.)
He took this attitude towards life into his philosophy for studying for the AP Latin exam. He bravely attempted to brute-force memorize every single line of Latin that we had learned that year.
On the other hand, I reviewed the general plot points and themes of each reading, a process which took about 2 hours, versus his 13-hour cramming session.
To be fair, neither of us got a perfect score on the exam.
And neither of us studied the way our Latin teacher wanted, which probably would have been a tidy combination of the two.
But ultimately, he and I both got what we wanted: he went to an Ivy League school, and I somehow got a full scholarship for Classics.
My methods may be unorthodox (… or pedagogically unsound,) but their results have everyone coming back for more.
Introducing: my friend Ellie, who is getting her Ph.D. in Philosophy and, for some reason, wanted my advice
Maia’s suggestion was to gauge how frequently the orators use certain keywords by using Perseus. By identifying those specific keywords and word frequency, distinguishing them from the “keywords” of other ancient Greek orators, I could use them as a tool (of course, not the only tool) for identifying who wrote which passages on my final.
Using the Perseus Vocabulary Tool for word frequency
I started with the Perseus Vocabulary Tool and clicked on the relevant texts for each orator we would be tested on.
The Perseus Vocabulary Tool is a tool for searching for occurrences of Greek and Latin words. When you select a text on the drop-down and then “weighted frequency,” the Tool outputs a list of words based on their weighted frequency within the text.
Selecting multiple (or all) texts from a given author will give you an idea of the words you will see more commonly in reading that author.
Why do I need weighted frequency?
In this context, “weighted frequency” refers to the prevalence of words in a text. It also accounts for the ambiguity of those words. In ancient Greek, single words take on various forms derived from different base words, making it challenging to sift through texts quickly.
“Weighted frequency” thus measures each word based on the number of “base words” it could be derived from to estimate its overall frequency. It’s much better than counting the number of times a particular word appears, as this does not account for the highly inflected nature of the Greek language.
Given this, sorting by “weighted frequency” would produce the most accurate results for determining the number of times a given word is found within a particular corpus.
Go to the Perseus Vocabulary Tool and select the texts you would like to study.
For instance, Demosthenes:
After I clicked “Show Vocabulary,” Perseus showed me the following screen.
I then copy-pasted the list above into ChatGPT, asking it to sort the list into nouns, verbs, propositions, and so on…
Input:
Could you take this list and organize it into the following: noun, verb, pronoun, preposition, etc.?
ὁ
οὗτος
εἰμί
δέ
αὐτός
μέν
καί
σύ
τίς
ὅς
ἐγώ
γάρ
τις
ποιέω
πρός
γίγνομαι
ἀνήρ
οὐδείς
περί
ἑαυτοῦ
ἄλλος
πολύς
ἐκ
ἐάν
ἐκεῖνος
εἰς
ὑμός
τοίνυν
πόλις
εἶμι
λαμβάνω
ἅπας
οὖν
βούλομαι
ὅστις
δίδωμι
πράσσω
ἀλλά
ὑπέρ
τοιοῦτος
ἐπί
ἐπί
ἀνά
ἄν
γε
ἔχω
φημί
γράφω
πρᾶγμα
ἐν
οὕτως
δικαστής
δίκαιος
μηδείς
μής
οἶδα
ὑπό
Output:
It appears that your list consists of Ancient Greek words. Here’s a rough categorization according to their common use, although some words can be used in multiple ways depending on the context:
Nouns:
ἀνήρ (man)
πόλις (city)
πρᾶγμα (thing, matter, affair)
δικαστής (judge)
Verbs:
εἰμί (I am)
ποιέω (I do, I make)
γίγνομαι (I become)
λαμβάνω (I take, receive)
βούλομαι (I want, wish)
δίδωμι (I give)
πράσσω (I do, perform)
ἔχω (I have)
φημί (I say)
γράφω (I write)
εἶμι (I go)
οἶδα (I know)
Pronouns:
ὁ (he, it – definite article often used as a pronoun in Greek)
οὗτος (this)
αὐτός (he, she, it)
σύ (you)
τίς (who)
ὅς (who, which)
ἐγώ (I)
τις (someone, anyone)
οὐδείς (no one)
ἑαυτοῦ (himself, herself, itself)
ἄλλος (other)
ἐκεῖνος (that)
ὑμός (your)
ὅστις (whoever, anyone who)
τοιοῦτος (such)
Adverbs:
καί (and, also)
μέν (indeed, truly – often used in contrastive pairs with δέ)
γάρ (for, because)
οὖν (therefore)
οὕτως (thus, so)
ἄν (possibly, might – used with a subjunctive verb to form a potential condition)
Prepositions:
δέ (but)
πρός (towards)
περί (around, about)
ἐκ (out of)
εἰς (into)
ἐπί (on, upon)
ὑπέρ (over, above)
ἀνά (up, upon)
ὑπό (under, by)
Conjunction:
ἐάν (if)
ἀλλά (but)
Particle:
γε (at least, certainly)
Adjective:
δίκαιος (just, right)
Interrogative/Indefinite pronoun:
μηδείς (no one)
μής (not) – this is often used in questions or negative statements
What do I do with the word frequency analysis?
I repeated this process with each of the Attic orators we’d studied this semester – Demosthenes, Isocrates, Lysias, and Gorgias – and pasted all of my information onto a Notion page:
Then, I color-coded different parts of speech and eliminated words from the lists that I knew wouldn’t be helpful (e.g., commonly used articles such as ὁ). I also marked in red the words that seemed unique to each orator. Having done this, I determined which Attic orators were more likely to use certain words than others.
For instance, following this method suggested that Isocrates often uses ὥστε (“so that”) in his speeches. This makes a lot of sense – Isocrates was well-known for his usage of parallel structure, antithesis, and long and artfully constructed causal and natural result clauses, many of which were marked by the particle ὥστε.
I analyzed themes.
Many of the terms frequently used by Attic orators also directly relate to the themes they explore. Demosthenes’ speeches, for instance, often revolved around the threat posed to Athens by Phillip II of Macedon. They were marked by force and a direct appeal to the people of Athens, and we see this in his recurrent usage of the vocative and the nouns πόλις (city) and ἁνήρ (man).
Interestingly, Demosthenes was the only orator we’d studied this semester who made frequent usage of the verb γράφω (I write). Demosthenes was the first (if not the only) great Attic orator to write his speeches down beforehand. I wonder if this frequent usage of γράφω has something to do with Demosthenes’ inclination toward writing down his speeches in advance.
Applying the above-described method, I also noticed an interesting correlation between the themes heavily explored in Gorgias’s speeches and the unique words that frequently occur in those speeches – words such as λόγος (reason), ψυχή (soul), δόξα (opinion), and πείθω (“I persuade” or “persuasion”). Gorgias was heavily concerned with exploring the nature of persuasion and reason and their interplay in his speeches. Gorgias aimed in his speeches to subvert conventionally accepted beliefs and make them appear absurd, thus questioning the nature of traditionally accepted truths, λόγος, and making an opinion, δόξα, appear to be true. Along these lines, Gorgias’s speeches also held a deeply moral psychologistic aspect within them – Gorgias was concerned with exploring the nature of the soul, ψυχή, and its susceptibility to being swayed by λόγος, δόξα, or persuasion (πείθω).
Back to Maia now
Recognizing rhetoric and literary devices is essential in developing media literacy. Implementing the practice in the Classics classroom, especially K-12 makes sense. Many students won’t be able to look closely at literature unless enrolled in an AP Language and Composition course. Sure, there’s overlap between the students who might take AP Lang and those who take AP Latin, but students should be exposed to this skill as much as possible.
The assignment that facilitated this learning quite well was the analytical essay in the AP Latin curriculum.
Fama assignment here on page 60
Comparing two seemingly disparate authors: one prose (Caesar) and one poetry (Vergil) can be daunting for students. It’s also a daunting task for teachers, given how much time must be dedicated to looking for passages with enough thematic similarity or overlapping vocabulary. Check out my discussion of the proposed AP Latin curriculum changes here.
However, this task can be significantly sped up with Ellie’s demonstrated technique. Teachers can quickly find the overlapping language or search for themes between texts to generate more examples of these analytical essay assignments.
Try it out for yourself! Let us know how it goes in the comments below, and if you enjoyed reading, don’t forget to subscribe for more.