How do we make sense of words in text? Letter by letter, whole-word recognition, or via sentence context? All of the above, it seems….

Coming to the final part of our series on FL literacy, today we’re going to discuss the possible complications in learning alphabetic writing systems.

First, it is necessary to note that, by “alphabetic writing systems,” I don’t only mean the Roman alphabet.  Many world languages employ alphabets of varying complexity — the Greek alphabet, Cyrillic (think Russian), Thai, Hebrew, Arabic, etc. all use very different characters, but they are all alphabets.  The common feature in all alphabets is that they use characters or character combinations to represent the speech sounds in the language.  As I mentioned last week, language depth — the regularity of correspondence between spelling and pronunciation is variable between languages.  Some languages, such as Spanish and Turkish, have a high correlation between sound and spelling.  You don’t often hear Spanish speakers having to ask how to spell each others’ surnames.  In fact, my students in Spain were amazed that English speakers have to do this.  If you hear a word in Spanish, you can typically spell it.  In English, words like “gnat,” “straight,” “laugh,” etc. are spelled rather counter-intuitively to say the least.  This is where you’re going to get the most trouble in the study of alphabetic languages.  All languages (including English, amazingly) that employ alphabets actually have very regular, systematic rules for spelling, but some simply have a much more complex set of rules.  In the case of English, a lot of the chaos of spelling is because of combinatorial rules necessitated by the fact that English has so many more sounds than characters (42-46 phonemes, depending on dialect, with only 26 characters to represent them).  Think of the fact that most Americans use 17 separate vowel sounds, with only 5 letters to represent them — it’s no wonder the spelling is a bit complicated.  Languages with a smaller sound inventory (like Spanish) or a higher number of characters (like Thai) have the luxury of better sound to spelling correlation.  Usually, going from a complex L1 to a more simply-represented L2 won’t be too problematic, but going in the opposite direction is highly frustrating.  Languages like Spanish, with the high sound-to-spelling correlation, lead to reading strategies much resembling the “spell it out” strategies employed by speakers of syllabary languages.  They must be taught explicitly to view the whole word before beginning any pronunciation, in order to prevent errors derived from letter combinations embedded in or at the end of words: ex. “photograph” vs. “photographer” or “bat” vs. “bate.”

There is also a minor issue of mechanical relearning, but this is, to not mince words, somewhat obvious.  Arabic and Hebrew are written from right to left.  A new alphabet like Cyrillic must be studied and practiced in order to draw the characters  in a recognizable fashion.  The conventions of writing vary between languages.   Any time you actually change the alphabet (i.e. when studying, say Greek), there is some “lag time” in reading processing.  We essentially are demoted back to kindergarten, as we need to concentrate on decoding sound to the detriment of our ability to decode meaning.  Nonetheless, most alphabets can be picked up rather quickly as long as one is diligent in study…

Today, we continue with last week’s thread on attaining FL literacy.  Last week, we looked at the three basic types of writing systems in the world, and we looked at some of the literacy-learning strategies used by native speakers of logographic writing systems (such as Chinese) that English speakers should be aware of in order to efficiently learn those languages.  Today, we will turn our attention to syllabic writing systems (syllabaries).
As one would expect from the title, syllabic writing systems have characters to represent syllables, as opposed to alphabets which represent  phonemes.  To explain the distinction, take the word “ma.”  In English, such a sound is split into two parts: /m/ and /a/.  In a syllabic writing system, such as the hiragana script of Japanese, a sole character is used to represent the combination of the consonant and vowel:　ま.  This is, of course, a highly efficient form of writing, but is only practical in languages with relatively simple syllable structures, and few possible sounds.  Japanese, for instance, only has 5 vowel sounds, which combine with 14 basic consonant sounds (k, s, t, n, h, m, y, r, w, g, z, d, b, p) and several dipthong categories (yo-on).  The only possible syllable-final sound is /n/.  The language can be represented by the hiragana script with a mere 46 basic characters, plus 25 character variants (reflecting voiced/unvoiced variations, such as “ka” and “ga”).  English, on the other hand, would never be representable in a syllabic format.  Think of the vast array of possible single syllable words in English — tree, think, top, train, stress, etc. — and you’ll understand why this type of writing is constrained by the language’s sound inventory.  Some other languages that have used syllabic scripts are Sanskrit, Ancient Persian, Cherokee, Nagari, Maya, etc.

An example of syllabic writing from Japanese

おげんきですか。   (How are you?)

  o  ge   n    ki    de su  ka

o = honorific (you)       desu = copula         genki = health               ka = question marker

In contrast to logograms, syllabic characters carry only phonological information (i.e., the characters themselves have no inherent meaning).  This means that, at least at a basic level, words must be sounded out in order to be understood.  In other words, the semantic meaning of the word can only be attained by accessing the phonology (sound) of the word.

Now, to get to the meat of today’s question: how do natives of languages with syllabic writing systems? While eye movement studies have shown that fluent readers can often recognize words without decoding each “letter” in sequence, low-level learners learn to recognize phonological information and to “string together” sounds into word structure. Syllabic languages have an almost absolute character to syllable correlation, not requiring the combinatorial rules familiar to alphabetic languages.  Thus, with few exceptions, low level learners can literally “spell out” text (though, just by way of being thorough, in Japanese, three characters can be pronounced differently when used as grammatical “particles” : は(ha)、を(wo)、へ(he) are pronounced “wa,” “o,” and “e” respectively when used as subject, object, and directional markers, respectively).  

For Japanese learners of English, this will cause some trouble in being able to isolate individual phonemes, resulting in the familiar “I supiiku Engurishu” (I speak English) accent.  For English learners of Japanese (and other languages written in syllabary), one would suspect that syllabaries would be relatively easy.  The one trouble to be aware of is that syllabary languages often have a sizeable distance between written and spoken pronunications, which can also be reflected in formal speech.  In standard spoken Japanese, for instance, word ending /u/ sounds are often dropped, however, in reading and/or formal speech, such sounds will be more clearly articulated, often resulting in decreased understanding for the non-native speaker.  Read text will also often have a robotic, syllable-by-syllable hum to it, in contrast to the normal word breaks and pauses of daily speech.  It’s not too difficult to get the hang of it once one is exposed to it, but it bares noting that reading patterns in syllabaries follow different norms than reading patterns in alphabets.

As noted yesterday, literacy learning strategies vary between languages. The highest amount of variation is invariably to be found between languages with different orthographic systems. Today, we’ll start looking at the major categories of writing systems in the world, and how to deal with them.

Logographs… Logographic systems are the oldest form of human writing… Technically, they are the second-oldest… Pictographs are older, but are often omitted as they are are not a “complete” system of writing — i.e., one can make a picture of dogs barking, but it would be difficult to express “the dog’s passionate howling at the loss of his owner evoked a sentiment normally reserved for a Mozart’s Requiem Mass in D minor.” Logographic forms of writing manage to alleviate that limitation by using homophones (words that sound alike) to stand for more abstract ideas (i.e., how do you draw “love”). In logographic writing systems, characters carry a meaning — not a sound, and thus — due to the incredible volume of vocabulary inherent to human languages — tend to have a very high volume of characters.

While Chinese characters are the best known logographic language, other writing systems have existed through history: e.g. Egyptian hieroglyphics and Mesopotamian cuneiforms. Actually, to be honest, we use logographs all the time in the U.S. For example:

While we all know intuitively what these signs mean, you’d be hard pressed to pronounce them. Chinese characters, of course are pronounced, but the pronunciation is arbitrary as there is nothing inherent in the character to indicate sound. They merely represent an idea. The following example means Chinese (person), and will show you some of the variation in pronunciation possible as the same characters are used by different languages.

中 国 人 (apologies to those whose OS doesn’t support Chinese script)

Zhong guo ren (Mandarin)

Diong gok lang (Taiwanese)

Jung kok yen (Cantonese)

Chuu goku jin (Japanese)

So, how do you go about learning a language like this?

Well, native speakers of Chinese have some distinct learning tendencies that it is important to be aware of:

1. Native Chinese speakers tend to learn their reading and writing skills through rote, word-by-word memorization, and frequent repetition (Chan, 1999).
2. L1 speakers of Chinese exhibit a large reliance on visual information in word decoding strategies (Chikamatsu, 1997). Children learning Chinese demonstrate a greater eye for minute detail than their English-learning peers.
3. Chinese-speakers also focus on semantic recognition of characters, as opposed to phonology (Pine, Huang, & Song, 2003). Characters have meanings, they do not carry fixed pronunciations.
4. Shu and Anderson (1997) determined that learners of Chinese made extensive use of knowledge of character radicals for determining semantic information. Literate readers also made use of radicals for recognizing less commonly used characters.
5. Highly literate speakers can also make use of phonological information sometimes embedded in characters, however, Shu and Anderson (1997) found this to be little used by lower-level learners – in native speakers, this strategy is not observed in a consistent manner until sixth grade (by which point most average children would meet government standards of basic literacy).
6. Kinesthetic recall methods, whereby a learner traces a character with the finger in the air, or with a pencil about an inch above the paper, are commonly employed (Pine, Huang, & Song, 2003).

The key for the learner of Chinese (or other ancient logographic languages) is to recognize that 1) you need to concentrate on visual details to distinguish characters; and 2) make use of semantic radical clues. Most teachers don’t teach this in so many words, but if you bring it to their attention, I’m sure they can help you to duplicate the tried and true methods of Chinese children.

Postscript: One final note — it’s important to realize that for Chinese-speakers learning English will have the same problem in reverse. It is often necessary to explicitly teach them to search for the phonological information inherent to alphabetic systems, lest they go about memorizing words based upon construction and shape.

When I first began studying Japanese, I was amazed at how much longer it was taking to learn compared to the other languages I had studied before. Likewise, when I started Chinese, even though already having a significant background in the Japanese language was a large advantage, and the oral/aural component came much more quickly and easily than Japanese had, I was still disturbed by the somewhat slow pace of acquisition. When I began analyzing how I studied Chinese, with how I studied, say, French, the culprit quickly emerged — whereas, when I lived in France, I was regularly immersing myself in print literature, my poor grasp of Chinese characters precluded like activity in Taiwan. I was thus left with the question: why did it take me so much longer to gain basic literacy in Chinese (I got better!) than it did in other languages? At first glance, it seems like an obvious answer — the Chinese language has a separate character for each word, so achieving literacy in Chinese is a daunting task, to say the least. To read a Chinese newspaper, one must learn approximately 2,500 characters (this is usually accomplished by the 6th grade by native speakers). The average adult has a minimal vocabulary of 7,000 to 9,000 characters — enough to read general books. Of course, people in more academic fields will usually have an even more expansive vocabulary. This alone seems to readily explain the difficulty of Chinese study, however, as I dug into the topic, I discovered a little not-too-well-known-or-discussed fact of language learning: the native speakers of every language have slightly different literacy learning methods, and an L2 learner of the language will be delayed if he/she employs different learning techniques.

Once stated, this makes some intuitive sense: it figures that native speakers, with the collective wisdom of hundreds or thousands of years of learning and teaching the script of their own language, would come up with the most efficient means of learning literacy in the language; however if not explicitly stated, many (most?) learners will automatically employ the literacy learning strategies used in their L1. Unfortunately, depending upon the regular similarity or difference of the L1 and L2, the L1 literacy learning strategies may work very little, if at all. By and large, teachers tend to be unaware of this. Most native speaker instructors will tend to phrase instruction in terms of how L1 speakers learn to read/write the language, but without explicitly noting the inherent differences between the two languages, many students will fail to pick up that, in addition to the difference between the languages themselves, the learning strategies used to acquire the written language must be changed significantly.

This is, of course, a lot more noticeable with highly differing (to English, at least) languages such as Chinese or Japanese; but what about relatively “close” languages such as Spanish or German? Even these will require somewhat different focus. Worthy of particularly close attention is the concept of language depth (the extent to which a language’s writing system accurately represents its phonology). For example, a Spanish speaker used to close correlation between pronunciation and spelling will be thrown by a language like English where to, too, and two are all pronounced like “tu.”

Much more details to come in the coming days — I’ll break it down into the major writing system groups, and tell you what to look for in each…

One of the first literacy issues that the FL learner will face is determining when to start studying reading and writing in the language, and what proportion of your study time should be spent towards literacy acquisition. If you enroll in FL classes, you are quite likely to start all four skills — listening, speaking, reading, and writing — on your very first day.

The first thing that need be realized is that this approach is, according to most research on the topic, misguided at best. Look at how children learn to read and write their first language. We don’t shove a book and pencil at them the day they are born (or even the second week). Children typically learn quite a bit of oral language — in fact they become downright expressive — long before they start reading and writing in earnest. Sure, people read to/with their kids at even the earliest of ages, but the child is still interacting with what the parent says — not what is on the page. (This is not to knock reading to children — it’s, in fact, quite vital in helping children to later recognize the relationship between print and spoken language). We allow children the time and experience to build their vocabulary, develop the rudiments of syntactic structure, and develop semantic and/or background knowledge before expecting them to interact with the written word.

Foreign language students, on the other hand, are typically not given enough time to develop vocabulary and grammatical knowledge before being immersed in print (though we can reasonably assume that most school-aged or adult learners will have adequate background knowledge). This is the approach that would work best for most FL learners as well. Laufer (1997) showed in a study that the size of a reader’s active vocabulary is the key to literacy. One requires a threshold vocabulary of about 5000 lexical items before one can avail oneself of such L2 literacy strategies as guessing unknown words from the context. In an ideal world, FL learners would be able to spend at least a year building their vocabulary before encountering print.

Encountering print too early can lead to frustrations due to the students’ inability to directly apply L1 strategies to L2 decoding (more on that tomorrow), but early on, one of the main effects is to distort phonology. If print is encountered before students have a good grasp of the FL sounds, the writing may serve to help solidify bad pronunciation and make it harder to break bad pronunciation habits. In the case of languages that share alphabets, this becomes apparent when low level students consistently apply L1 sounds to L2 words (such as using the “a” sound from “apple” in the Spanish word “manzana”). Even with unfamiliar scripts, the ability to transcribe words makes students less dependent upon and inclined towards careful listening to the foreign language sounds.

This is not meant to be an indictment of the public school system. Formal FL classes have good reason for introducing text so quickly. While I would bet that most teachers are probably aware at some level that it would be beneficial to delay the introduction of the printed words until students had time to develop functional vocabularies and the ability to actually use the language in real conversation, most foreign language programs are simply pressed for time. The average American student will only spend two years studying a foreign language, and it would shortchange the student significantly to be deprived of learning literacy skills in the language during that time. The demands of assessment (i.e., testing) also necessitates the rapid introduction of reading and writing. Thus, formal language classrooms will not be spending the first 1-3 years in an all-oral environment any time soon. I will end, however, by merely encouraging anyone who is planning on making a longterm commitment to their language to investigate the possibility of adapting your study plan to emphasize oral skills and vocabulary building during at least your first year. Have something to read and write about before you attempt reading and writing.

Getting to this post late on a Monday. Sorry about the tardiness, but I had a hectic day…

This week I wanted to look a bit at the issues and possible trouble-spots in becoming literate in your foreign language. While, let’s face it, for most of us, the most pressing need will be oral comprehension, it’s really good to be able to read as well. For some learners, such as those learning their FL for research purposes, reading may be the main goal — but it must be noted that this is a minority. Many languages don’t even have writing systems — or at least what is in existence is mainly used by linguists and missionaries (for Bible translation), and not used much by the indigenous population. That being said, when a writing system is in wide use in a population, the ability to access that gives innumerable benefits, among which are: 1) better access to information; 2) the ability to effectively learn aspects of the language itself by oneself (i.e., looking in a dictionary); 3) the ability to gain information and learning from people not here in the present time and place; 4) etc.

That all being said, it is important to note that gaining literacy of any type (let alone FL literacy) is actually a more daunting endeavor than one would typically imagine. Humans are not endowed with an innate ability to read and write, as we are endowed with an ability to learn to speak and understand spoken language. Any able-bodied child given regular input will learn to speak and understand language, but reading and writing have to be explicitly taught. Most of us also spend very little time interacting with the written word, as opposed to the spoken word. It’s usually estimated that the average person spends up to 70% of the total time spent in communicative acts during the typical day merely listening to someone else. Speaking time accounts for another 20+%. Reading usually accounts for only 5-10% of total interaction for the day (although, I feel compelled to point out that for grad students like myself, that total is often much higher). Most people can easily go days at a time without writing anything (us internet blowhards bloggers excepted). Of course there will be personal variations on the totals — these are mere averages — but we can see that most people don’t interact with writing nearly as much as they interact with speech.

Thus merely becoming literate (in any language) requires more effort than most of us really realize. It’s not easy to remember how difficult something is after you’ve already mastered it, and considering that most of us started reading and writing out native language at age 6 or younger (my mom was apparently teaching me to read when I was two — I can’t claim to remember that at all, but I’m told they were good times). Learning to read and write in a foreign language, however, has its own difficulties, which we will discuss all this week….