"Tenses" (as in past, present and future). Part 1

I’ll soon be writing some morsels on the GPA Facebook page on the cognitive dimension of the GPA. Suffice it to say, for the GPA, “grammar” is not “instructions for how to form sentences” but part of a “cue system” which is used by the “comprehension system” in the process of converting sound into understanding.

“Past-present-future tense” is a powerful theme in our Anglo-world folk-theory of language and language learning. When people want to do a “modified GPA,” one of the common concerns will be, “The GPA doesn’t teach past, present and future in Phase 1, and that’s really basic”. If you ask, you’ll find we have something to say about every area in which there are common complaints/misunderstanding! So this message explains some of the reasoning behind the way “tense” is dealt with in the GPA.

First a bit of terminology: Tense, Aspect, Mood (TAM)

1)   I am writing this blog entry

2)   I was writing this blog entry

3)   I wrote this blog entry

Sentences 2) and 3) are said to be past tense. They differ in aspect. Sentences 1) and 2) are progressive aspect (which is a type of imperfective aspect). I hope that gives you a bit of an idea of the basic distinction between tense and aspect. When it comes to mood or major concern is with imperative and interrogative, but we won’t dwell on mood now.

A problem in talking about TAM is that all our examples are in English, which uses such categories in particular combinations, for particular purposes. As we are really talking about many languacultures, we need to be more general, and so we talk of “tense,” meaning tense and aspect. Sentence 1) is here-and-now “tense” (or we might say, here-and-now tense-aspect). Sentence 3) has what we will call story-event “tense” and sentence 2) has what we will call story-background “tense”.

Note that we put "tense" in quotation marks. The realities of languages will differ greatly. For example, in the imperative there might be a contrast, which we don’t have in English, between imperfective (“Be writing a blog!”) and perfective (“Write a blog!”). Some language may not have tense, that is, using a particular aspect for the story-event form. Some languages have hodiernal tense (past, but still today), pre-hesternal (past, and prior to yesterday), etc. Lots of different aspects, too, in the languages of the world. Huge variety, and all sorts of complexities.  The “quotation marks” around "tense" are there to remind you that I’m using the term in a vague, everyday way that reflects our Anglophone-world folk-theory of language and language learning.

It seems that in the SLA (second language acquisition) field, few researchers get beyond thinking of “tense” as a way of referring to a “location in time”. A common idea is that initially “language learners” express time by adverbs, such as “yesterday,” before they learn to refer to “locations” in time by tense marking.

In my dissertation, I observed the oddity of practice, by lingusits (influenced by a certain logician),  of viewing tense/aspect marking as a way of expressing the fact that the “Event Time” preceeds vs. coincides vs. follows the “Speech Act Time”. If someone is telling a typical story, then all the events in the story precede the speech act times. So why mark that fact on every single verb. Yet historically, a language without past tense marking will develop it, perhaps over several centuries (this is called grammaticisation). It seems to be doing something more important than (totally redundantly) reminding us that each event in the story happened before the time when the story is being told.

Some linguists, such as Paul Hopper and Sandra Thompson have proposed that “past tense” (etc.) marking has a different function. It marks the “foreground” events—that is the events that move the story forward. Some psychologists have understood this concept of moving the story forward in terms of “mental models”. As we hear a story, we have a mental model that keeps changing, developing in line with the events of the story. Daniel Morrow had people listen to two versions of a story. One had the words, “John walked through the kitchen into the bedroom.” The other had the words, “John was walking through the kitchen into the bedroom.” These differ in terms of aspect, perfective (walked) versus progressive (was walking). The effect of the “simple past” (the perfective) in the first case is a mental model in which John is in the bedroom now. The effect of the progressive is a mental model in which John is now in the living room. The effect of tense/aspect then is to move us to a particular place in the story line.

Now whereas it makes no sense that languages would have a great tendency to mark the time of every event in a story (prior to the Speech Act Time, etc.) when that is so obvious, it makes excellent sense that the tense/aspect morphology would help people understand where we are in the story.

I would argue that in general the “time reference” of tense marking is incidental. For example, “future tense” is in fact marking speech act types such as making a commitment to do something. Of course, whatever you commit to do, you will do it in the future.

To be continued…

Name that sound! [GP or nurturer shivers.] Right! (Or wrong!)

Get it? The name of the sound was--in your new sign language--a shivering action. (I get a chill just thinking of it.)

Well, this seems to be the blog for more cognitive topics. If you want a more clearly sociocultural flavour (at the moment), better turn to the Facebook page. However, the way we sound when we talk is, in fact, a highly important aspect of our identity to host people. Each of us will always have an accent (not just in speaking, but right across the board, in our hearing, our understanding of words, the inferences we draw, the grammatical form of our utterances) which will be such a part of who each of us will be.

I remember as a nineteen-year-old thinking that articulatory phonetics was the magic bullet for quickly achieving accent-free speech. Decades later, I studied speech perception, and found out that matters were more complicated than my articulatory phonetics teachers realised. You don't get to hear whatever you want hear. Rather, your mother-tongue listening system strongly influences what you hear when you listen to host speech. Granted, host speech does "clear up" a lot over the early days and weeks, and pronunciation does improve a lot over the first six months or a year.

Language courses that try (in vain) to be exhaustive with all the articulatory phonetics are not in the spirit of the GPA. Too much time is spent talking about "retroflex voiceless aspirated stops" and on and on and on. The GPA tries to minimise the intellectualisation/academicisation of growing participation.

It is in the GPA spirit to give a name to "retroflex voiceless aspirated stops", either a familiar host word that starts with a "retroflex voiceless aspirated stop," for example, ThanDa, 'cold', or a gesture representing that word, such as a shivering motion with one's arms torso and head. In this manner, the nurturer and GPs come to have a set of shared names for troublesome sounds--increasing the closeness of their personal relationship, unlike "retroflex voiceless aspirated stops," which increases the distance in their impersonal relationship!

Now the GP hears a new word, say, Thosna 'stuff, cram', and to confirm that she heard it right, she makes that shivering motion. Or if the GP badly mispronounces a word, the nurturer makes the shivering motion, to tell the GP which sound she is supposed to aim for. Rather than shivering, the GP or nurturer could also say the word for 'cold' aloud, if that is the chosen name of the sound in question. (In choosing the names of sounds, use words that the GP is already familiar with from earlier sessions.)

Can you see how this way of categorising and referring to sounds is more, well, friendly?

However, as a GP, don't try to exhaustively name every sound. Concentrate especially on sounds that you have great difficulty discriminating. Sound distinctions that you readily hear can gradually tune up your pronunciation. However, if you can't hear a sound contrast at all, then how will you know whether your own speech is conforming to host speech when it comes to that sound? You may learn to produce a contrast that you can hear, based on some articulatory instructions. However, if you can hear the contrast you are producing, but  you can't hear the host contrast, then the contrast that you are producing is not the host contrast!

Starting late in Phase 1A, give a little attention each day to hearing host sound contrasts, using sound sorting, or listen and point with words that are almost identical.

For most people, their best attempts at describing the articulation of sounds ignores complex mechanisms of both hearing and articulating. An example would be the "explanation" that, "In aspirated stops, there is a puff of air following the release". Well, that is absolutely true, but it misses most of what is happening. If the vocal folds are closely approximated, then the airflow will be reduced, and will cause the vocal folds to vibrate. By contrast, if the vocal folds are far apart, then the airflow will be greater than when they are closely approximated. The longer the speaker waits for the vocal folds to come together (for example 20 milliseconds, 50 milliseconds), the larger the airflow will be. The waiting time before bringing the vocal folds together, even if the sound is said to be "unaspirated" or "aspirated" in articulatory phonetics, differs from language to language. Now even if you explain this complex mechanism (and there would be more to explain indeed), the GP doesn't have conscious control over actions like waiting 10 milliseconds and then drawing the vocal folds close together. So you'd be better off telling a GP, "it's a sound half way between a 'b' and a 'p'."

Furthermore, you'd only mention that if the person is majorly mispronouncing it. And then you wouldn't torture the person trying to get him or her to pronounce it perfectly before going on. We improve over time, and mainly because our hearing improves, and our pronunciation can then conform better to our improved hearing.

In Phase 1A, Session 1, Game 1, your hearing is vague and "blurry" (for people of most mother tongues hearing most host languages). You distinguish words based on salient features, such as "starts with 'pa...', or "has a strong 'mu' syllable in the middle somewhere". Quickly the "neighbourhood density" of words grows. There are now, two words which start with "pa..." or have a strong "mu" syllable in the middle somewhere. "Neighbourhood density" refers to the number of words that are similar to a given word. Once you know thousands of words, the neighbourhood density of many will be great, and you'll be forced to register a lot more detail than you did in Phase 1A, Session 1, Game 1! This is all part of the process of host pronunciation "coming clear".

After 30 silent hours in Phase 1, you start talking. Now your struggles to pronounce those familiar words will also move you to a new level in your listening for phonetic detail (and, we would claim, will have a much better impact on your listening than had you started talking in Phase 1A Session 1, Game 1).

Your own speech becomes clearer over the months as you interact intensively, and gradually, gradually, manage to make yourself more and more intelligible to host people through trying and trying. You are drawn to sound like the people you talk to.

With all this in mind, it just doesn't make sense to teach every GP a supposedly exhaustive description of every sound contrast using fancy latinate terms. Not a good use of time. It misleads them into thinking more is possible than is, fills their minds with "science" rather than filling their heart with nurturing, and it limits the amount of time that can be given to strategically learning to discriminate contrasts they truly have trouble with.

Therefore, GPs should especially concentrate on distinctions to which they are "deaf" or partially "deaf", and perhaps a few others that really seem egregious to the nurturer. But remember that which particular sound contrasts are difficult for a GP depends on the GP's mother tongue. So it would be wildly inefficient to have a detailed enough phonetics course for everyone regardless of mother tongue. Help with pronunciation needs to be personalised, and given with patience, and not overdone at any point in time. It is well known that GPs' pronunciation can be highly intelligible even though strongly accented. GPs will be strange to host people in so many ways, and pronunciation will be one of the more obvious ones. Teachers complain that phonetics lessons just don't seem to help a lot of people (perhaps citing particular nationalities) as though it is the people's fault. No, it is the teacher's fault for having an overly simplistic view of the challenges of developing host-like hearing and pronunciation.

Don't go overboard. Better to go underboard! Hearing and pronunciation keep improving for many months. If you intervene in trying to help a person to pay more attention to a sound contrast, intervene strategically, and be prepared to accept partial or total defeat. People with strong accents nevertheless become dear!

Why not stop calling Arab languages/varieties "dialect"?

Recently I was in the Middle East and noticed this jargony thing among expats of saying “dialect” (in the singular) in contradistinction to “MSA," for example,  "MSA vs. dialect"; "Fusha vs. dialect". At least people could use a capital "D" for "Dialect", since it is conventional in English to capitalise the names of languages. However, in different cases in the Arabophone world, the specific language referred to as “dialect” will differ. But it would be better to actually name the language or variety, such as Khaleeji, Hassaniya, etc. rather than calling them all "dialect" in the singular, as some sort of mass noun. Saying things like “MSA vs. dialect” suggests a sort of balanced binarity which contributes to a distorted understanding of Arabic diglossia. It feeds the mistaken idea that MSA is “correct” Arabic, while the normal Arab vernaculars are just degenerate forms of MSA.

Anyway, it demeans these Arab speech varieties to call them “dialect” with a small “d” as though there is one reality across all of the Arabophone languacultural worlds that is “dialect” in various instantiations. Of course, the role of MSA in pan-Arab unity is wonderful. However, there is no need to demean the specific Arab speech varieties by which people live their lives–which they hear first in the womb, then are born into the midst of, raised in, skin their knees in, cry in, get comforted in, be playmates in,  laugh in, love in, be friends in, be families in, grow old in, tease grandchildren in, die in, etc. The GPA is about people nurturing us into their life. Human life is dominated by talking, listening, interacting verbally. 

Within a languacultural world, there may be more restricted functions which rely on special language varieties (such as MSA). To the extent that the use of the special variety finds its place among host practices, its use are part of the host languacultural world that the GP wants to be nurtured into, but keeping the realities of the time dimension always in view. Once people can talk with you readily, they can, by talking with you, nurture you into literacy, literature, style-shifting, etc. 

So while it is wonderful that MSA has a uniting function, and has strong historical links with Classical Arabic (and is thus a bridge to Classical Arabic) and so on, and certainly, MSA should be treasured as much as it is, this does not require that the speech varieties by which Arabs primarily live life should be demeaned by calling them all “dialect".

More on the "Here-and-Now-Photos-of-Us" game

I've talked a bit about how we abandoned the "Here-and-Now-Descriptions-of-Us" game in Phase 1, and replaced it with the "Here-and-Now-Photos-of Us" game. The latter game was actually describe in a couple of places in the Phase 1 guide (The First 100 Hours), but we've elevated it. Below is an explanation I wrote for another forum.

The old-cowboy traditional-linguist in me naturally thinks of person and number categories as first base. (person&number means I, you, you-plural, we, he, she, they, and various other possibilities). But now the Vygotskyan LLA in me also feels that the starting point of all languacultural life is two people in relationship, against the backdrop of others, and making reference to those others (and inheriting much from them). In other words, the kernel of languacultural life is "I-you" plus "he". It grows from there.

In other words, nothing seems more basic than person&number. Given this conviction, I was bothered by the fact that "I-you" played so little role in Phase 1A. Well, we get off to a good start in Session 1, Game 1, with "man, woman, I, you, boy, we..., dog, cat" (Where is the man? Where am I? Where is the dog? Where are you? etc.). Rip roaring start. However, think of a new baby: "I" and "you" are going to be some of the most frequent bits of sound in his/her environment. We may agree that they are the kernel, but they don't seem to play much natural role for us until dialogues in Phase 2. So we came up with the "here-and-now-descriptions-of-us," but always felt they were a lame patch up. In that game, you may recall, everyone starts doing something--sitting, walking, running, etc. And the nurturer addresses a single GP in the group, telling that person what everyone, including the nurturer, is doing, "*I* am standing. *You* are walking. *They* [two of your fellow GPs, say] are running. Now *we* (you and I) are sitting and *he* is walking... " 

What was the task in this? Well, the only task is to notice how the nurturer says those things. Not too good. Angela pointed out that it is in the spirit of "point and listen" which we largely shun, and not in the spirit of "listen and point," which we praise. (In "point and listen" there is no choice to be made, and so we don't process what we hear in the same way as when we have to do the work of  understanding it.) In the case of Kazakh, we videoed this game, which meant at least we could relieve and refresh the experience and strengthen the forms. 

Before entering the Kazakh world, at the same stage of Russian growing participation, we did something better. The year was 1995 or 6. We took our camera to our session, and we made separate photos of different combinations of us doing things separately and together. We only met once a week with that nurturer, and so we had a week to get the photos developed (we went to a one-hour place--the wonder of technology). 

In our next meeting, we spread the photos around, and the nurturer could literally ask, "In which picture am *I* running," since there was indeed a picture in which she was running. She said to my ten-year-old, "In which picture are *we* walking." He kept choosing the wrong photographs. Finally, she realised he could not understand "we". So she pointed to herself and said, "This is I". She pointed to him and said, "This is you." Finally, she put her arm around his shoulders and gave him a very firm sideways hug, saying, "This is we". Now that, my brothers and sisters, epitomises sociocultural learning of "we". First person plural is a good spot to epitomise that, isn't it. Few concepts are more sociocultural (or more GPA-ish) than "we," after all.

This game with photos, as an alternative to "here-and-now-descriptions-of-us," was presented in the Phase 1 guide, and I've mentioned it at many LLACs. However, we never named it. Now we have: "Here-and-now Photos of Us". Think of how superior it is to the mere descriptions. It creates a real information gap. The GP must process the person&number forms and understand them and use the information to perform that task, which means much deeper processing than when you are just supposed to listen and understand. It is also a good example of a "structured input" game. It can hopefully be modified for any level of complexity of person-number-gender-etc. system. (But if the languaculture uses both pronouns and other person markers, don't forget to omit the pronouns for structured input--or else you won't listen to the other person markings. So instead of "I am running; he is walking" you would just hear "am running; is walking".) 

Remember that with structured input, you need to keep it simple, as in "Start with two concepts, and only add one new one at a time". Also, and I'll put this in all caps, since they say that ALL CAPS IS THE EMAIL VERSION OF SHOUTING: DON'T GET CARRIED AWAY AND LET STRUCTURED INPUT ACTIVITIES TAKE OVER. They should be a really small part of the whole picture. And stick to grammar issues that are relevant to here-and-now language (in other words, not past tense and future tense!).

Having given this warning, I can testify that when done in proper balance, such games are powerful. In Potwari for the present progressive you have a sequence of a person, number and gender suffixes and mixed with other suffixes/particles. A little bit of structured input for several days--and at first even that seemed overwhelming--but by the time we started talking, we could produce the forms ourselves (not that it would have been the end of the world had we not been able too--remember, it's a long road).

So take a felt-tipped marker and go through your Phase 1 guide (The First 100 Hours) and wherever it says "here-and-now-descriptions-of-us" cross it out, and put "Here-and-now-Photos of Us". If you start Phase 1 again yourself or are coaching a Phase 1 group snap the required photos of the specific group on your phone or digital camera, and print them on your inkjet printer, and do the activity, a little at a time, as long as it is helping. With Mandarin Chinese, that will be for about five minutes! For Blackfoot, it may be helpful for several months. (Angela says Gulf Arabic looks a lot easier than Potwari.)