English-Speaking Hackers Should Learn About Esperanto

by Robert L. Read.  Copyright Robert L. Read, 2003, 2004.  You may freely distribute this document if you do not change it.


If you are a computer programmer with the hacker attitude, by which I mean valuing competency, it is a worth spending a few hours studying Esperanto. There are a lot of reasons one might want to learn Esperanto; I am going to focus on why the hacker should learn at least a little about Esperanto.

Esperanto is a language invented in 1887 to foster world harmony as a culturally neutral universal second language.   Such constructed languages are referred to as conlangs on the web, although the term planned language might be better, as Esperanto is a living and growing language. It by far the most used conlang, having at least a few hundred thousand speakers and perhaps one million or more.  You can purchase about 300 (40% of them original) novels in Esperanto from the Universal Esperanto Assocation (UEA), although the English speaker might prefer the catalog of the Esperanto League of North America or simply to examine the very large collection of free works available.  At the time of this writing (Nov, 2004), Esperanto is the 14th most commonly used language in the Wikipedia (behind Danish and Hebrew, but ahead of Finnish and Norwegian).  The best resource for learning Esperanto is lernu.net.  A good book for learning about Esperanto is The Esperanto Book by Don Harlow.

Many persons think of the idea of a language that is designed rather than evolved as "artificial" and unattractive in the same way that plastic plants are unattractive.  The programmer, to whom nothing is more natural than the purposeful construction of a language to ease communication, will, I trust, be intrigued rather than repulsed by this feature.  Although there are other conlangs, such as Elvish, Klingon, Ido, and Lojban, they are essentially unused compared to Esperanto.  Esperanto therefore holds a singular place in being a constructed language that is alive, widely used, and offer the same benefits as learning a national language.  (I have heard Modern Hebrew described as so different from Biblical Hebrew and so recent as to be a conlang; in that case, Esperanto would have a partner.)  I personally want to study  Spanish, Arabic, Lojban, Mandarin, French, Klingon, Irish, Japanese, Elvish, and Russian (in that order) --- ha!  To me, and I think to the intellectually curious everywhere, the question is in not whether or not we should study a given language, but in what order, understanding full well that we are likely to die before getting very far down the list.

I would therefore like to examine the benefits and costs of studying a language, and in particular the advantages and disadvantages of conlangs and national languages, and Esperanto in particular.  (I waited until I obtained fluency in Esperanto to write this essay; I am not talking completely out of my hat.)

Roughly speaking, studying a language offers two benefits:  it is an exercise for the mind, and it lets us communicate more.  I cannot prove to you that studying a language makes you smarter, but I think it does, in the same way that studying chess makes you smarter.  The effects of such mental exercise are not overwhelming; playing chess won't make you witty, and learning a language probably won't make you much better at math.  However, I personally suspect based on anecdotal evidence that mental skills are fairly similar to physical skills in that practice helps, and to some extent one skill bleeds over into others.  I wish we had a good model for mental athleticism, that would allow us predict the benefit of a certain kind of mental exercise.  Lacking such, we can rely on one widely reported anecdote:  the more languages you learn, the easier it is to learn the next language.  Here we have a virtuous circle.  This affect is sometimes called the propaedeutic effect.  One language is a prop for the next.

This effect presumably works in proportion to the proximity relationship between language families.  That is, within families, languages are similar, at all levels: phonetic, morphemic, syntactic (I'm probably leaving out an -ic.)  For example, French, Spanish, Italian, Portuguese and Romanian are all in the same family (Romance) and fairly close.  Learning one is a diret and demonstrable benefit in learning the next.  German, Hindi, the Romance languages, the Slavic languages (Russian, Czech, Polish), and Greek are all siblings, if you will, in the group of Indo-European languages.  Presumably, learning these far less self-similar languages benefits less in learning another, but still there is some benefit from each language learned.  Arabic and Hebrew are Semitic, and not Indo-European.  The East Asian languages and the Indo-European are foreign to each other.  However, there is still a benefit in having learned an unrelated language before beginning one---a person learns how to learn with practice.

If you take a typical American who knows zero langagues beyond English, one can roughly describe the difficulty of learning a langage about like this:


Language
 Difficulty Factor
Esperanto
 1
Spanish
 4
Klingon
 7
French
 8
Elvish
 11
Arabic
 12
Mandarin
 20

Obviously, at one level this is silly---infants learn all languages equally well.   (It is often said that infants and children learn languages quickly.  There is some truth with respect to the phonological processing in this.  However, I personally am very skeptical of this effect.  I think it is over-emphasized and used as an excuse by many people.  If an adult could be relieved of all responsibilities, put in an immersive environment, and have their ability to obtain food be based on speaking with extremely devoted caregivers fluent in a language, I supect the adults would learn faster than the children.  This would be in keeping with the propaedeutic effect, in any case.)

However, at another level, it is true.  English prepares you pretty well for Spanish because there are a lot of cognates, especially to the well-educated English speaker.  Spanish is also very regular, and completely phonetic, so it is relatively easy.  I'm sure many would disagree with the above table, but more will generally agree once they have picked the nits off it.  Notice that I am asserting Esperanto is four time easier than Spanish.  Since I am fluent in Esperanto and have studied Spanish four years in high school with some trips to Mexico and a little study on the side, and have presented, badly, a technical paper in Spanish, I can assert this from personal experience.  I may have been a little unfair to Klingon and Elvish; since Elvish is still an incomplete language at this writing and Klingon doesn't have many speakers.  I have started studying Arabic enough to have personal knowledge of how foreign and difficult it is for an English speaker.

So obviously, in considering to study a language, we should at least know the difficulty we expect, since, after all, the main cost in learning a language will be time.  Of course, one will spend some money on books and dictionaries and software and so on; but anyone who is actully using those valuable materials will spend more time-money, even at minimum wage, working with them than purchasing them.

It may be hard to believe Esperanto is that easy.  It is that easy.  It was designed to be easy, and whatever else one may say about it, it succeeded splendidly in that respect.  Briefly, it is easier because:
Esperanto is a very logical language, by which I mean, to a human, it is logical how the language functions.  There is a conlang called Lojban which is designed to represent logic; my impression is that it is logical at the expense of its ease.  This may be an advantage, but that advantage is different than the one that Esperanto offers: extreme ease of use and learning.

Dearest Reader, I know your time is valuable.  You may not put learning Esperanto high on your list, but please considering spending a few hours learning about Esperanto.  You will hear the programmer in you, and perhaps the language learner, sigh "Ahhhhhhh!" as you appreciate its simplicity and power.  

I have asserted that the cost of learning Esperanto is low.  This is, of course, relative.  Trying to actually speak and use Esperanto for me was difficult.  It took me two years of about 30 minutes a day before I could claim intermediate fluency in Esperanto.  However, I can put this in direct relation to my experience with Spanish; Spanish is a lot harder.  (I  still hope to become fluent in Spanish.  In fact, I am hoping that having mastered Esperanto will make it relatively easy when I return to it, in a few years.)  However, Esperanto is so much easier that one can see one's progress very clearly, and this is, again, a virtuous circle.  I think it took about four months before I read my first horror story in Esperanto, the translation of a gem by H.P. Lovecraft, Pickman's Model, and gave myself the willies.

Let us now consider the benefits of learning a language.  The direct commercial economic benefit of learning a conlang (even Esperanto) is approximately zero.  The economic benefit of learning a national language depends on one's particular circumstance, but roughly speaking we can be hard-nosed and use a basic formula to compare languages.  I propose something along the lines of: (the number of speakers) times (the per capita income of the speakers).  This formula suggests why many people are worldwide are learning English, why Mandarin is a challenger to English, and why English may someday fall off its pedestal as Latin, Arabic and French have more or less fallen out of international ascendancy.   To an American, if you look strictly at economic benefit, it is hard to justify studying anything other than Mandarin, especially if you project into the future.  (For everyone who is not an English speaker, you should learn English for the same econmic reasons.)  Of course, there are many benefits beyond the strictly economic.  For example, Esperanto is widely known as a match-making language.  Arabic and Japanese give you access to centuries of largely untranslated literature.  Learning a less popular language, like Irish, is highly appreciated by the speakers of that language, since they know you derive no economic benefit from it.


However, setting aside acts of love, let us remember that Mandarin is, perhaps, twenty times harder to learn than Esperanto.  Sure, there are ~800 million mandarin speakers, and only ~1.2 million Esperanto speakers.  Of course, Esperanto speakers are highly pre-selected for high education level, and therefore high incomes.  Esperantists are friendly, chatty, high-brow, well-educated, free-thinking, and they throw excellent parties.  But putting all that aside, its hard to argue with 800 million speakers of rising income in terms of the economic benefit of knowing a language.

However, let us remember the propaedeutic effect.  If learning Esperanto speeds your learning of Mandarin by 1/20th, you would be well served to study Esperanto first, as it would barely postpone the date of your Mandarin fluency.  It is hard to imagine why any English speaker setting out to learn Mandarin would not spend a few weeks or months studying Esperanto first.  Since one can be a converstationalist in Esperanto in just a few months, during those years of studying Mandarin, you could write to and perhaps visit the many Mandarin speaking Esperantists.  Esperantists, being part of an active, quixotic movement, are highly likely to be interesting, friendly, and supportive of other Esperanto learers.  They even have a world-wide network of homes in which you can stay for free, subject to varying restrictions, if you speak Esperanto.  They are only unreasonably likely to be male if they come from North America.  It seems likely, therefore, that using Esperanto as a warm-up before starting the Mandarin marathon might help you cross the finish line sooner, rather than later.  To me, Esperanto is like the techniques used in weightlifting.  One is ill-advised to exert maximal effort by lifting one's absolute maximum liftable weight as one's first, or only, exercise.  To me, Mandarin seems to be a maximum weight, and Esperanto a good strengthener.

Of course, learning French first would also help you learn Mandarin faster, and you might value knowing it more highly than knowing Esperanto.  However, it still seems reasonable to learn Esperanto first, then French, then Mandarin.  I wish we had a predictive model of such things.  It would be wonderful to be able to say, "Ahh, I see you have good Spanish, and high-school German, and enough Japanese to be cheated trading Yu-gi-oh cards.  Three-hundred hours of Esperanto will cut your French fluency time by 250 hours.  If you put 1000 hours into French after that, Mandarin will take you 6000 hours, instead of the 10,000 for your average American."

I remember distinctly being in junior high and attempting to read a mathematical paper.  It was menacing and confusing, because I did not know what the symbols meant.  I remember there was the symbol we use for set membership, and I thought it was an oddly typed letter-e.   The expression "it's all Greek to me" applied to me in this case.  What I didn't know, and would not learn for another seven or eight years, was that a significant hurdle to any mathematics is learning the sub-language that it uses.  There is, of course, more to math than just the parts that are similar to learning a new language.  And of course, learning Esperanto will not directly teach you the sub-languages used by mathematicians.  However, I think there is something to be said for the flexibility of mind and self-confidence that comes from having mastered a foreign language helping the student of math, or some other discipline that uses, and in fact demands, a strange language.  To me, a page filled with formulae is as equally terrifying and glamorous as a page filled with Arabic.  

Learning a national language might get you a job, and will expand your mind. I encourage you to do it---after you have studied Esperanto.   Learning Klingon might expand (or at least twist) your mind.  I encourage you to study conlangs other than Esperanto---after you have studied Esperanto.  To do otherwise would be to study painting with the attitude that zero knowledge of drawing is needed.  I encourage you to construct new languages---after you have studied Esperanto.  The programmer must, if he is to be effective, consider the construction of new languages.  Obviously, these are programming languages and embedded languages and data representation languages and data manipulation and other dry things that you can't make love with.  However, once you have studied Esperanto you will understand why I think I can legitimately speak of natural (human) languages and computer languages in the same paragraph.  There is in all design work the principles of simplicity, concision, completeness, naturalness.  Esperanto has minor linguistic flaws, by which I mean that almost everyone who learns it can think of some minor way in which it might, at first glance, be improved.  But time has proven it to be a jewel of design, that any hacker can appreciate.  J.R.R. Tolkein certainly did.

In short, I think the benefits are so great and the costs are so low that studying Esperanto, at least a little, would be a smart use of time for any English-speaking with an attitude of intellectual curiosity.