Contents
How Not To Learn a Language
I hold the distinction of having learned classical Greek, Latin and Hebrew (to varying degrees), forgotten them all, and re-learned them. In school, my teachers used the conventional “Grammar-Translation” method. I memorized all the verb forms (conjugations) and noun forms (declensions), structures for various clauses (purpose, result, etc.) and basic vocabulary. Once all that was memorized, then I read the assigned text, dutifully looking up each word in the dictionary and analyzing the grammar of each sentence to come up with an English translation. I got excellent grades in college, appearing on the Dean’s list multiple times. However, by the end of my undergraduate degree I was frustrated and discouraged. I could not pick up a Greek text and just read it for pleasure. I was told that to reach the level of fluency I desired would require a graduate degree. I had spent enormous effort on my Greek degree and had almost nothing to show for it other than a piece of paper to remind me of the four years I wasted. For this and other reasons, I abandoned my life-long goal of becoming a Greek scholar and instead pursued a career in technology.
Throughout the years of my technology career, my interest in languages (especially biblical languages) persisted. I still wanted to enjoy reading a book in another language for pleasure. I dabbled with French, Russian and Spanish but never made any significant progress. My mentality was captive to the methodology my teachers used. I thought that before I could read another language, I had to put in lots time of memorizing forms and grammar and vocabulary by rote. Only then would I have earned the right to read for pleasure. In short, Grammar-Translation failed me. It’s tedious, boring, difficult, slavish labor. Worse, aside from being anything but fun, it is largely ineffective because only an exceptional few who use this approach succeed in learning the language at a level even remotely close to fluency.
A Far Better Approach
Eventually, I came across the book Lingua Latina Per Se Illustrata which relies on the concept of Comprehensible Input. This book is an beginner text for Latin. The entire book is written in Latin and yet somehow the reader (who doesn’t know any Latin!) is supposed to be able to understand it. The way the reader learns to read Latin is by…reading Latin! What a concept! The book presents each new word with a labeled picture illustrating the meaning. Then, beginning with the shortest possible sentences repeating the word from multiple angles, it incrementally builds up to more complex sentences which introduce grammar concepts and word forms in a comprehensible context. The book is a joy to read.
Multiple organizations now offer language classes using the Comprehensible Input approach which is essentially immersion. My objective here is not to reproduce the voluminous content freely available on the Internet about how to learn languages most effectively. Rather, I simply want to state explicitly that my approach to learning Hebrew is guided by the principle of Comprehensible Input and reading to learn (instead of learning to read).
Correct Pronunciation is Essential
When my professors spoke Greek words, they used a conventional pronunciation known as Erasmian, so named after Erasmus of Rotterdam who formulated it. My teachers and textbooks never told me this, but Erasmus himself acknowledged that the pronunciation was not entirely correct, i.e. ancient Greeks didn’t actually speak that way. The attitude was that it doesn’t really matter if the way we pronounce Greek now would be recognizable by ancient Greek speakers. All that matters is that we are relatively consistent.
As it turns out, pronouncing the target language correctly has a major impact on one’s ability to read and understand the language. Sloppy pronunciation makes reading and understanding the target language surprisingly difficult. There is academic research on this topic if you are so inclined to dive into the details. Suffice it to say, I took a short course on correct classical Greek pronunciation (Attic Greek specifically as opposed to multiple other ancient dialects) to fix my horrible Erasmian pronunciation. Changing my pronunciation dramatically improved my ability to read Greek.
The same is true for biblical Hebrew, which is slightly different from modern Hebrew. So, the very first skill to acquire in order to read Hebrew is to pronounce it correctly.
Shadow Reading
The central technique I use for learning a language is to imitate a native speaker as they read a text. Shadow reading refers to listening to a native speaker and imitating their sounds, intonation and phrasing. This assumes I have access to a native speaker reading the text I’m interested in. Fortunately, there is a freely available audio recording of the entire Hebrew Bible made by native Hebrew speakers. Actually, there are several sources of recordings, but the best by far is provided by the Bible Society of Israel’s web site Haktuvim (“the writings”).
This web site has everything needed to shadow read effectively and efficiently:
- Parallel English and Hebrew text
- Mouse over a Hebrew or English word highlights the corresponding word in the parallel column
- Audio embedded on the page with the text
- Audio controls that enable skipping reverse and forward in 10 second intervals
- Ability to play the audio at multiple speeds, most importantly slower (0.5 or 0.75 normal)
The procedure for shadow reading with Haktuvim is simple:
- Search for the desired Bible passage.
- Read the sentence in English to have a mental idea of the meaning.
- The audio recordings are segmented by chapters, so if the desired passage is inside a chapter, use the audio navigation controls to locate the starting point in the audio. When beginning to shadow read, it is best to start at the beginning of a chapter.
- Press play to start the audio. If the default normal speed is too fast, select a slower playback speed.
- Say out loud the sounds the reader makes in sync with the audio. Same pitch. Same intonation. Same accent. Same emphasis. Same pace. Same everything. Imitate exactly. Pretend you are a native speaker.
- At first, shadowing the reader will be difficult. Set the playback speed to slow and click “-10” on the audio controls to skip back 10 seconds to repeat the phrase. Continue skipping back and attempting to imitate the reader until you can say it perfectly. Expect that this will require many, many repetitions: 10, 20, 50, 100. However many it takes. It takes time (likely multiple sessions over a few days) to learn the physical skill of controlling your tongue and shaping your mouth to produce the unfamiliar sounds in an unfamiliar order. It will feel like a tongue twister at first.
- Continue this procedure word by word, phrase by phrase. Do not proceed to the next word or phrase until you can imitate the previous words perfectly. If you proceed before you are ready, you won’t acquire the skill and you will simply handicap your ability to read. Resist that temptation. Slow becomes smooth, smooth becomes fast.
- After you can shadow read a phrase fairly well, think about what each word means as you read it. Don’t attempt to translate it, simply imagine the meaning. For example, if the Hebrew word means “ship”, don’t translate it by thinking the English word “ship”, rather visualize what a ship looks like. As you continue reading repeatedly, the meaning will get connected in your brain to the Hebrew sounds without understanding any grammar (just as children learn).
- Celebrate the feeling of immense satisfaction when you finally experience imitating the native reader correctly and fluidly, and you also understand what those sounds mean.
Maximum Time Efficiency and Vocabulary for Free
Suppose I only have 5 minutes to spend shadow reading. What can I accomplish in 5 minutes? Assume the audio recording of the verse / sentence / phrase is 10 seconds long. If I shadow read that phrase repeatedly without stopping, in one minute I will hear and say the phrase 6 times. In five minutes, 30 times. There is an extremely high probability that after repeating the phrase 30 times, I will be able to say it correctly. Not only that, my brain has associated the sounds with the meaning 30 times. In effect, I put all my effort into pronouncing the sounds correctly and I get the vocabulary for free, no extra effort.
Target Highest Frequency Words First
The Hebrew Bible contains over 9,000 unique words [see here]. The top 600 most frequently occurring words account for approximately 85% of all occurrences. So, learning the 600 most frequent words will cover the vast majority of vocabulary I will encounter.
It would be convenient if most of those 600 words appeared in a one place, and it turns out there is such a place: the book of Jonah. Jonah contains approximately 980 total words and 250 unique words [see here]. Most of those 250 unique words are among the 600 most frequent words. So, shadow reading Jonah not only builds proficiency in pronunciation but also provides more than a third of the most frequent words in the entire Hebrew Bible.
The same is true for the book of Ruth. This is why Jonah and Ruth are typically the first books intermediate Hebrew students read. Not only that, but Jonah and Ruth are great stories. There’s emotion and drama, danger at sea and a monster, loss and a love story. They are fun to read over and over.
CoPilot Knows Hebrew
Generative AI is a boon for language learning. Multiple freely available AI services can be used to answer specific language questions. The one I have found most helpful is Microsoft CoPilot. If I have a question about a Hebrew word or phrase, I copy/paste the Hebrew into my prompt and CoPilot provides the answer. Prior to the widespread availability of generative AI, I would have to dig deep through books hoping I could find the answer. With AI, I have the equivalent of a expert in the language at my disposal anytime I need. And I can freely ask questions that I would hesitate to ask a human because the question seems embarrassingly simple or I already asked but forgot the answer.
For example, I can prompt CoPilot with “The first Hebrew word in Jonah 1:1 begins with a vav. Is that conjunctive or consecutive?”. CoPilot responds with “The initial ו on וַיְהִי in Jonah 1:1 is vav‑consecutive, not merely conjunctive.” Without CoPilot, I would have to find a book or other source that provides the exact parsing of the word. One such freely available source is this doctoral thesis. Addendum A “Morphological Analysis of the Book of Jonah” (page 377 of the PDF) provides the parsing of every word in the book. A textbook that contains the answer is A Workbook for Intermediate Hebrew by Robert Chisholm Jr.
Generative AI occasionally produces wrong answers. So, I critically think about every answer given, check sources and take nothing for granted. It’s a powerful tool, but I don’t trust it without verifying.
Long-Term Perspective
Learning any language well is a long-term endeavor. What reason or purpose justifies all the time and effort? The only justification that makes sense to me is that I intend to read Hebrew, Greek and Latin for the rest of my life. This is a skill I want. Specifically, I find great value in reading the Bible in its original languages. Furthermore, there is a veritable mountain of extra-biblical literature written in these languages. In fact, the quantity of extant literature in these languages is far greater than I could read in multiple lifetimes. The practice of continually reading keeps these languages fresh in my mind. Put another way, if I don’t plan on using what I learn, I will lose it and the investment will be wasted. “The secret to your future is hidden in your daily routine.” I have a clear vision of what I want my future to be (fluently reading Hebrew, Greek and Latin), so it is part of my daily routine.
Next: Learn the Hebrew Alphabet