My methods won’t work for everyone, but I hope some of it is useful to you. Get a basic grasp of the grammar before you start memorizing vocabulary. Lojban words are easy to memorize, but learning to correctly structure your Lojban text requires effort. It is not just a substitution code for English. The way the words work together is unfamiliar and different from any natural language.

Resources

I recommend you start with the quick Lojban Guide written by the founders. I put together a nice document of it’s most relevant opening section. (400KB PDF) Then use the exhaustive “The Complete Lojban Language” or the lighter “What Is Lojban?” Those are links to the online texts, but you can also buy nice paper books of CLL and WIL.

If you use the Firefox web browser, it can serve as a super-fast Lojban compendium at your fingertips. Go to the Bookmarks menu and select “Organize Bookmarks”. Create a folder called Lojban and drag it into the Bookmarks Toolbar Folder. Exit “Organize Bookmarks”. Drag the following links into that folder’s button on the Bookmarks Toolbar:

tool to parse Lojban text for grammar correctness

tool to return English gloss words for Lojban text

interactive dictionary

list of all root words

list of all function words

list of all function words ordered by family

list of all prefix/suffix forms of the root words for use in compound words

big list of all root words and function words ordered by how common they are

Nora LaChevalier’s list of compound words

Lojban IRC chat on the web

The Complete Lojban Language by John Woldemar Cowan

Lojban wiki

Custom Lojban Google

Vocabulary Memorization

I started this process in August 2004 on the Supermemo flashcard program for the Palm OS. Check out that introduction to the method. Supermemo is available for many platforms. Use Robin Lee Powell’s personal Lojban page to get it set up with Lojban flashcards. Supermemo works so well that after I skipped a year, I could still remember most of them.

The reason I skipped a year is that my Palm OS device and the hard drive to which it was backed up crashed at the same time. Twice. The new models of Palm devices keep your data even when the battery runs out, and can back up to SD cards. Only when I got one of those did I resume flashcards, starting over from the beginning. Make sure your backups are secure.

At first you’ll have zero tests and drills because you haven’t committed any cards yet. To commit a card means you are ready to start on it. You’re not intended to commit the entire wordlist at once, because committing cards is the way to set your own pace of learning.

Your First Flashcards

Before you even start SuperMemo, you should understand the pronunciation rules, and understand what makes a gismu (root word) different from the parts of speech used in your native language. This knowledge is used in most cards.

Generally, commit new flashcards in the order the words are used. There is a Frequency field in the database you got from Robin’s site, and Supermemo will let you sort the whole database by that field.

However, if you’re not ready to use a word in a sentence, skip it for the time being, or you’re creating what’s called “memory interference.” This means it does more harm than good to create links in your brain between some noises and definitions that mean nothing to you. Go to the textbooks “The Complete Lojban Language” and “What Is Lojban?” to look up how a confusing word is used. Then, when you can use it in a sentence, commit the flashcard for it.

You’ll encounter many definitions referring to “bridi” and “sumti”. It’s a good idea to look up the flashcards with the definitions of those words and commit them first. If you’re using the standard databases you’ll have to create a flashcard for “selbri.” Do this when you find yourself ready in the textbooks to advance to certain classes of words whose definitions contain an unfamiliar Lojban word, such as “brivla” “gismu” “rafsi” “lujvo” “fu’ivla” “tanru” “cmavo” “prosumti” “probridi” “discursive” “abstractor” or other jargon. If there are no cards for them, make them.

Next, learn a common, simple gismu (a root word such as “cusku: express”) and some common, simple prosumti (”do: you”, “mi: me”) so you can make a simple utterance as you memorize. It’s helpful to speak a word in context of a complete simple sentence when you review it. That involves more cmavo (function words) than gismu (root words), because in order to speak effectively at all, you need the most basic cmavo a lot more urgently than any particular gismu. You can swap in all kinds of gismu into the same sentence structure, but you need some cmavo to create a sentence structure.

Regulating How Much Work Supermemo Gives You

For the first year or so, I used a method to even out how much work I wanted to do in a single day. I would set a target number of cards I was aiming to do each day: most months I set the number at thirty. Here’s how I calculated how many new flashcards I wanted to commit that day to avoid overworking myself.

# of new cards = (your target #, minus # tests in Lojban-to-English) + (your target #, - # tests in English-to-Lojban) / 2 rounded up

Let’s say, at the start of a particular day, Supermemo presented me with twenty-four tests in Lojban-To-English, and eighteen tests in English-To-Lojban.

I’d subtract each of them from my target number of thirty. Thirty minus twenty-four is six. Thirty minus eighteen is twelve.

I’d add the results. Twelve plus six is eighteen.

Then I’d divide that in half. Eighteen divided by two is nine.

So, I would commit nine new words for memorization that day. This method was intended to regulate Supermemo so I’d end up doing about thirty flashcards a day.

Memorization Beyond the First 2,000 Words

Although it will at least triple my number of vocabulary flashcards, I thought I might memorize the place structures by making individual flashcards for the meanings of the second, third, fourth and fifth place structures of each gismu (root word). For instance, “klama: go” would become “se klama: destination” “te klama: origin” “ve klama: route” “xe klama: vehicle” and so on. And of course, this would involve creating the reverse flashcards to test them from English to Lojban as well. But then xorxes suggested a way that would drastically reduce the number of flashcards:

 you don’t need to learn the place structure for each animal separately, you just learn that animals have the place structure “x1 is a XXXX of species x2″, and then you only need to learn the three exceptions. Also in some cases you can identify patterns in the exceptions. For example, body-parts generally have the place structure “x1 is a/the XXXX of x2″, but for some number of them it’s “x1 is a/the XXXX of a/the x2 of x3″, so you only need to learn which of the two classes of body-part the word belongs to. There are other exceptions too, but still it is easier to learn the exceptions than each word independently.

Cmavo Which I Don’t Intend To Learn

Now that I’m most of the way through memorization, only these rare and obscure function words remain. Understanding them, even in English, would require more math and formal logic than I have, so I am unlikely to ever use them. I question their immediate value to my practical fluency, so I’m focusing elsewhere for the time being.

Return to my main Lojban page.