Interested in language?

Here's the reason I haven't been updating my website or sending out my newsletter for a long time — I've been working on a dictionary. The Indo-European Cognate Dictionary, to be precise. It's out now, and I'm really excited about it. Excited that it's done, excited than I now have a physical copy that I can use myself, excited because — hey, I've written a dictionary! It weighs in at 545 pages, and it's available in the usual digital formats, except Kindle. If I work out how to get Kindle to display the non-Roman scripts properly, it will appear in that format too, but don't hold your breath. Why Kindle can't do this, when ePub (the format used by practically every other reader) does it no problem, without any special measures needed, escapes me totally, but there it is.

Anyway, what's this odd-sounding dictionary about, you may ask — and what does it have to do memory & learning? Well, if you've done any study of a language related to your own, you'll know about cognates. They're the words that are easiest to learn because they're similar to the words you already know, because they're descended from the same root word. Words like Spanish importante / English important; French authentique / English authentic; German Apfel / English apple; Dutch mixen / English mix; Iceland eyland / English island. How easy a language is to learn is partly down to the number of shared cognates — Spanish, for example, is one of the easiest languages for English speakers to learn, because there's a huge number of words that are very very similar.

But not all cognates are as obvious as the examples I just cited. For example, English ban is related to French fable, to Spanish infando, to German Konfession, to English microphone to Polish bajka to Hindi भाषा (bhāṣā) to Dutch bahasa. All of which looks interesting but not particularly helpful, until you contemplate the full set, what I call the cognate cluster:

*bʰeh2-

to speak, say

*bʰéh2-ti, *bʰ-né-h2-ti, *bʰeh2-meh2, *bʰéh2-os, *bʰoh2-neh2, *bʰeh2-ni-s, *bʰéh2-ti-s, *bʰh2-tó-s, *bʰéh2-tu-s, *bʰeh2-dʰlo-, bʰh2-new-ti

Germanic: ben, bene, bee, boon, ban (Eng); bēn, bannan (OE); banne (Fris); bannen (Dut); bannen (Ger); bón, bœn, banna (Norse); bønn, banne (Nor); bön, banna (Swe); bøn, bande (Dan); bón, bæn, banna (Ice)

Derivatives: bannir (Fr); banish (Eng)

Italic: fārī, effārī, effābilis, fandus, īnfandus, nefandus, fāns, īnfāns, fātum, fāma, īnfāmis, īnfāmia, fās, fastus, nefās, nefārius, nefāstus, fātus, fatērī, cōnfitērī, cōnfessiō, prōfitērī, prōfessiō, professor, fābula, fābella, fābulōsus, fābulātiō (Lat);

effable, enfant, fameux, infâme, infamie, faste, néfaste, confesser, confession, profession, professeur, fable, fabuleux, fabulation (Fr);

infante, fato, fama, infamia, infamare, fasto, nefasto, confessione, professione, professore, favola, fiaba, favella, favoloso (Ital);

infando, nefando, infante, hado, fama, infame, infamia, fasto, nefas, nefario, nefasto, confesar, confesión, profesión, profesor, habla, fábula, fabuloso (Sp);

infando, nefando, infante, fado, fama, infâmia, fasto, nefas, nefário, nefasto, confessar, confissão, profissão, professor, fábula, fala, fabela, fabuloso (Port);

faimă, profesiune, profesor, fabulos (Rom)

Derivatives: effable, ineffable, infandous, nefandous, infant, fate, fame, infamous, infamy, nefarious, nefast, confess, confession, profession, professor, fable, fabulous, confabulation (Eng);

faam (Dut); infam, Konfession (Ger); professor, fabel (Swe); fabúla, fabúlera (Ice);

инфант (infánt), конфе́ссия (konféssija), профессия (professija), профе́ссор (proféssor), фа́була (fábula) (Rus); fabuła (Pol); famë (Alb)

Hellenic: φημί (phēmí), φαμί (phamí), φήμη (phḗmē), φωνή (phōnḗ), φωνέω (phōnéō), φάσις (phásis), φάτις (phátis), φατός (phatós) (AnGk); φήμη (fími), φωνή (foní) (Gk)

Derivatives: -phone, phono-, antiphon, cacophony, euphonious, homophone, microphone, megaphone, phone, phoneme, phonetic, phonics, polyphony, symphony (Eng)

Slavic: ба́йка (bájka) (Rus); bajka (Pol); bajka (Cz); bájka (Slo); бајка (bajka) (Mace)

Indo-Iranian: भनति (bhánati), भाषा (bhāṣā) (Sans); भाषा (bhāṣā) (Hin)

Derivatives: bahasa (Dut)

If you study this cluster, you can start to see the patterns in the way the Proto-Indo-European root word has evolved and spread through its daughter languages. Once you realize that f (ph in Greek) and b are the same letter in a sense (that is, the b sound in the root stayed the same in some daughter languages, but changed to the similar f sound in others), then you can see similarities in the words that you missed before. Once you recognize the regular suffixes and inflections that modify words in Greek and Latin, that are then carried on down in their descendants, then you can see the core of the word, and understand what the modifying bits indicate.

And the reason why I thought this was really useful, is that learning the sheer number of words needed to be fluent, or even useful, in a language, is the bit where most people fall by the wayside. Now, as I've said before, the keyword mnemonic is really good for learning vocabulary, but you want to keep its use to a minimum. Any word you can remember easily by simpler means is a word you don't have to try and think of a mnemonic for.

Remember too, that learning requires retrieval practice — this is your main strategy. Strategies such as mnemonics are adjuncts. It's all about reducing the number of times you need to practice, and increasing the length of time you can remember for without practice. And understanding is better than artificial connections (which is what mnemonics is about).

So this cognate dictionary is for increasing the number of words you can learn through natural connections.

The dictionary covers 430 cognate clusters, involving 40,000 words in 32 Indo-European languages. These aren't all treated equally, however — I wanted to show the breadth, and get a foothold on the less popular languages (in terms of English native speakers learning other languages — this is not a value judgment!), but the bulk of the words are from the Germanic and Italic (Latin and the Romance languages) branches (of course, this also works in reverse — it's not just about English-language learners of other languages, but also for those learning English as a second language). Here's a breakdown of the numbers:

English: 6378

Latin: 4130

Spanish: 3089

Italian: 3000

Portuguese: 2933

French: 2720

Romanian: 1770

Ancient Greek: 1377

German: 1223

Dutch: 1081

Russian: 907

Old English: 895

Swedish: 892

Norse: 890

Sanskrit: 818

Icelandic: 776

Norwegian: 762

Danish: 743

West Frisian: 601

Modern Greek: 562

Polish: 549

Czech: 490

Lithuanian: 457

Welsh: 422

Slovak: 411

Irish: 389

Albanian: 370

Latvian: 316

Macedonian: 295

Persian: 245

Hindi: 175

Pashto: 69

The dictionary contains the cognate clusters (arranged thematically, and cross-referenced as needed), but also indices for each language, so you can look up a word in your chosen language to find its cognate cluster.

The book isn't just about learning languages, though. There was a lot of very tedious grunt-work involved in putting this together, and I couldn't have done it if I hadn't been sufficiently intrigued by the things I found. If you're interested in words, I think you'll find this dictionary quite intriguing. I certainly did!

By the way, I should note that I was inspired to produce this book when I started listening to the History of English podcast, which begins in those far-off Proto-Indo-European times and moves forward. Great podcast, and Kevin also has transcripts available at his website. Highly recommended!