Institute of Semantic Restructuring

Navigation

Semantic Restructuring is the pursuit of enlightenment, enlivenment, empowerment through the creative re-arranging of the building blocks of meaning. For a better description, Start Here.


2004:09:07

964 - Objective my...

I am curious about the growing reliance on google for statistical fact checks. More and more I'm seeing the best and the brightest using google searches to subtantiate their claims, as in this post at Language Log.

I recall, a good while back, when I was first learning about the vi/emacs holy war, there was the "EDITORs Sucks-Rules-O-Meter", which lead me to the Operating System SROM, which in turn lead me to the "Tool of Objective Truth" (formerly at zdnet uk, apparently the plug's been pulled on that one.) I can't imagine anyone ever thought the Sucks-Rules-O-Meter was actually measuring anything; rather, it counts reports. To say there are possible sampling issues with those reports is to put the matter mildly. And even if one were to argue that the sample is sufficiently controlled as to warrant drawing conclusions, the conclusions drawn would still be about the opinions of the universe of respondents sampled and in no way reflective of any objective, criteria based analysis of the tools in question (the kind of analysis that would show Windows can't suck as much as we linux lovers love to say; if it did there wouldn't be the steady flowing river of formerly-winDoze boxes on which to run linux.)(Put differently, what linux really rocks at is a.) making the most of hand-me-down machines, b.) implementing unix. The great majority of users don't give a good gosh darn about the OS; they just want their fancy typewriter/jukebox to work as advertised and not have to cope with any learning curve greater than that associated with setting the clock on a VHS recorder. For these users Linux is still a nightmare, Mac still a little obscure, and the gates virus rocks and rules supreme...however much that truth may turn our stomachs.)

Back to my concerns with using google as a representative sample of language use. The language refelcted in the gooble db is language filtered from the norm in some non-trivial ways. It is the language of literacy; try as we might to forget it, the language of writing is only superficially related to language in general; literacy is not a criterion for language use. As if this weren't enough of a skew toward elitism, this language of literacy is further filtered by the criteria of publication; even a publication in a blog betrays relevant socio-cultural memberships and status. Plus, and this is the one that most concerns me, how the heck do we know that the database itself is inviolate? Sure, there is no conceivable reason anyone would monkey with the data in the google db; the more accurately the google db reflects what the google spiders have found the more valuable that db is. Still, the assumption that there is no benefit to someone skewing the numbers isn't the same as conclusively establishing the validity of those numbers...validity, that is, of those numbers for the select sub-set of language use that those numbers represent.

Oh well; this kind of google searching is damned useful, a great start, better than nothing. I guess I'd just like to see the occasional disclaimer along the lines of, "The language represented by these google searches is skewed in favor of that dialect or social register associated with published writing, an elite subset of language use in general." The Operating System Sucks-Roles-O-Meter page has links to "Tool of Objective Truth" and "COMPLETELY UNSCIENTIFIC comparison of programming languages." Sadly, both of those links have fallen to link rot since my last visit, sometime in the 20th century. Still, even as busted links, they represent something simple but important: one of them tells it like it is, right up front.

Addendum
I wrote Geoffrey K. Pullum, the author of the relevant language log post, trying to get trackback turned on for said post, here is an excerpt of his reply:

I drew no statistical conclusions of the sort where sampling bias could arise; the numbers merely show that his claim is wildly, overwhelmingly wrong, a piece of unchecked hyperbole so extreme that it should never have been made...I don't see that the objections are worth raising: they are hypothetically relevant, to kinds of points that I didn't try to make.(Emphasis added: RL)

Entirely true and valid. I would have done well to add a disclaimer of my own that while I worry what casual readers of language log will make of such examples I am certain that the folks posting there are too sharp to miss such subtleties. That my little rant was such non-sequitor is plain embarassing.

[] static link
writebacks: 1 (writeback = trackback +/- comment)

965 - Tell Me How to Feel

Sometimes we need to be told how to feel.

This may seem to violate precepts of authenticity and transparency as preached by some pop-psych movements, but the reality is you already feel much of what you feel because you've been told to feel that way. And sometimes the way you've been told to feel is wrong. Sometimes what we need is for someone to tell us it's ok to feel something else.

The thing people forget, or overlook, or simply don't realize is that most of what passes as natural, spontaneous, intuitive feelings is still programmed, patterned, learned. "I know it like the back of my hand," is short hand for "I don't think about it." "As intuitive as 2 + 2 = 4," is a way to say, "I've edited out my personal struggles to learn numbers, arithmetic; they were so long ago I don't count them any more." Even "alphabetical order" is so ingrained in us that most of us never question it. Think about it. What order is there in the alphabet as we know it? Not shape. Not sound. Not size. No, the order of the alphabet, that we take so intuitively for granted, is learned, taught, reinforced a thousand and one times.

So too for our feelings. We are taught and trained to feel good about shiny new things. We are taught and trained to want to be skinny young things. We are taught and trained to feel bad if we don't have shiny new things and can't be skinny young things. There is no inherent reason to feel that way. There have been times in history when skinny young things were considered sickly and shiny new things were considered dangerous, they were things to feel bad about. That was then, this is now, and the fact is these feelings are learned, ingrained, constantly reinforced from repitition and socializing factors. It isn't as simple as your pappa sitting you down on his knee and saying, "Hot young tail, that's what life's all about." Instead we soak up these injunctions about what is supposed to make us feel good and what is supposed to make us feel bad from a constant stream of informal examples and cues; jokes, movies, stories, love songs, all the aspects of the culture.

And so there you are, September 1, 2004, and you feel however you feel about things. And there is a pattern to that. Not a random pattern. An arbitrary one, an arbitrary pattern shaped by the intersection of your culture's values and your personal experiences. Not random. And not truly intuitive, but learned, accreted over time. And sometimes you might notice you don't like the way you feel. Sometimes you might notice that the arbitrarily accreted pattern of feeling responses you have today just doesn't serve you as well as you would like. That's when we need someone to tell us how to feel.

Of course, my telling you to feel good about your belly doesn't carry much weight against the culture screaming at you that if you don't have six-pack abs your are unfit to be seen. That message will get reinforced every time you turn on MTV or run the guantlet of magazine covers prominently posted at point of sale at your grocery store. You are subjected to thousands of messages a day, not only telling you what to buy, but why to buy, how to feel about not having what you would have if you bought. Buy this record, be cool. Buy this car, be rich. Buy this exercise machine, be fit. And feel good. Unless you can't buy it, then feel bad. So my simply telling you that you are a creature of light, a divine spark, part of a divine flame that is all creation, well, my saying it directly once or twice doesn't carry as much weight as the constant barrage of messages telling you that unless you are Brittany Spears or Brad Pitt then you're nothing and no one and should be happy to have your pathetic job and pittance wages and miserable life. Because repitition counts.

Well, since repitition counts, I'll say it again. As you are, changing not a thing, you are a divine construct of a divine will as part of a divine plan to create beauty and wonder. Is that a little heavy for you? Don't worry, it might or might not be any more true than the lies Cosmo and GQ tell you about who you are and how you should feel. Chances are you long ago swallowed the lies of Cosmo and GQ. If it comes down to my lie or theirs which will you choose?

Islamic, Hindu, Buddhist, Christian, Taoist systems of thought share a common element, instructions for living a better life. And while there may be conflict as to what defines a better life, there is another thing all these systems share: practical exercises for entraining your feelings. Call it prayer, meditation, manipulating prana, accessing the protective unconscious, rearranging the building blocks of meaning, no matter; there are overlaps between all these approaches that are of great value when the task is to entrain our feelings out of the arbitrary patterns of the moment into more enlightened, lively, powerful patterns. So when I say it might be a lie that you are a divine spark of a divine flame, well, it might also be true. You won't find final answers along those lines on this site. But you will find constant reminders that you can be better to yourself, kinder and firmer, sweeter and wiser. You will find a constant reminder to feel better, about yourself, your prospects, your essential being, the ways of the world. Because sometimes we need someone to tell us how to feel.

[] static link
writebacks: 0 (writeback = trackback +/- comment)

966 - Dip a knife for the knife

There are abstractions of such vintage coin and ubiquitous use that they become foundational to a culture. Veneration of "science" would be an example of an abstraction going in this direction, even when comically misapplied, as in Christian Science or Religious Science (fairness disclosure: I was raised in a spin-off of the Church of Religious Science, and when I need the social comforts of a church here is where I go, although these days I lean more towards taoism's briliant, but lonely, "if you can talk about it you're not really talking about It.") Back to abstractions. Abstractions such as point-line-space, thought-idea-meaning, word-sentence-language have long since seeped into the bedrock below the concrete slab of the building that is Western culture; there is no building without these abstractions, and it's mighty hard to get at and analyze them wihout ripping down the building, smashing the concrete slab, mining the bedrock and carting it off to the lab. If only study of language were so straighforward, rather than being the infinitely more subtle, slippery complex task it really is.

A friend writes:

(Point, line, space) form a sort of shorthand that makes the universe thinkable, that allows us to make generalizations. They are creations of nature. They constitute a sort of "folk science," a pre-science without which our modern sciences would not be possible.

"Makes the universe thinkable" is a really nice, sweeping turn of phrase. But thought, meaning, universe are all foundational abstractions, in the woof and warp of the language and culture. If there is any one risk in linguistics it's this one: that the tool we use for our field of study is itself the field of study. It's awfully hard to use a tool and at the same time free oneself of all the assumptions built into the very shape of the tool. Sure, one *can* dip water with a knife, but it's not the best tool for the job. Likewise, it is possilbe, concievably, to use language to investigate language, but it's not even on a par with the knife/water example, it's more like when an infant reaches for its right hand with its right hand, like the knife dipping not for water but for the knife.

Probably the most concrete example I have of this kind of Whorfian influence of language on world view comes from programming languages. At the simplest, it is possible to write an infinite loop (and cause a machine to crash) simply by moving from one language to another and forgetting how they each handle equality (that's personal experience, btw, from my first week on the job as a web programmer.) Call that first order world-view-shaping, as in knowing the difference between "the same word" in two different languages (i.e., two noise systems that sufficiently map the same things the same way such that a simple change of noise will work, as in English "day" and German "tag.") More delicately (and I might call call this second order world-view-shaping) there is at least one programming language, perl, that declines to differentiate whether a variable is a string of letters or a number or even another variable. The perl compiler looks at the operator and uses the contents of the variables to suit; so, for example, if you see $x == $y, perl treats the contents of the variables as numbers; if you see $x eq $y perl treats the contents of the variables as letters. Many other languages insist not only that a varialbe be assigned a type, they also then diffentiate between different kinds of strings and numbers (strings usually are a matter of largest allowed size; numbers tend to diffentiate size and method of calculation: integer, non-integer...) So when programming in perl one is spared making many distinctions that other languages insist on. Yet many of the languages that insist on detailed variable declaration don't insist on case-sensitivity and shun the use of characters outside the alphabet (whereas perl is rife with *@#%& and insists" this" is not the same as "THIS" or "This" or even "tHis", a pair of two edged issues that fans hail and critics hate.) These differences approach the kind of stuff that orginally captured Whorf's imagination (if I've read my history right) that the Hopi language simply had no direct analogs for anything resembling the Western, linear conceptualization of time, much as many programming languages have no equivalent of a simple perl scalar. It is easy to get a little nervous when talking about Whorf, but I think the baby has been tossed out with the wash water. Only a fool would say, "Perception governs reality;" sadly, many have, and used their misunderstanding of Whorf to support their claim. But, similarly, only a fool would deny that one's perceptions greatly influence our interactions with reality. And language is certainly interdependent with, influencing and influenced by our perceptions.

To tie back in, the programming language one uses will strongly affect the way one goes about solving problems or acheiving results (perhaps two ways to say the same thing but reflecting affectively disparate world views, in turn resulting in effectively different worlds) becuase the language of choice simply works differently, is better at doing different things, built to different criteria.

Natural language is first and foremost, it seems to me, accreted from primate noise and gesture systems in conjunction with changes in brain structure to solve the problems and acheive the results common to primates (which, at the risk of letting my [Freudian] slip show are eating, mating, and spacing {with spacing not only an abstraction of access to food and mates but something more integral to the organism.})

Time is movement, and the smallest quantum of time contains movement, as anyone who owns a camera knows.

I maintain time isn't even movement, but rather an inferred entity. I likewise hold space to be an inferred entitiy. In general the argument goes: Zero is an abstraction for the absence of countables, it is qualitatively a different animal from the natural numbers. Likewise space-in-which-objects-reside; there is no sensory evidence of this space stuff, it has no empircal existence. So too for time. But it is almost imposible to talk about anything without recourse to these concepts, and the tool of discourse fails even to permit a means of using these concepts cynically, reservedly, advisedly; to speak at all, at least in English, is to assume at a very deep and slippery level that Time, Space and other such non-emprical entities exist. To say time is movement is to conflate what we observe with what we infer about the observation; empircally we can say P moved from M to N. Empirically we can say that we observed O move from Q to R. Empirically we can describe the one in relation to the other. So far there is no need for time. But I don't know how to talk about the relationship of these two points and their respective travels without using words like "as, while, during, when" all of which presuppose time. The best I can do is plot both line segments and say "where x=3 P=(x3,y1) on line segment MN and O=(x3,y2) on line segment QR," but although I have avoided using constructions that presuppose time I've also terribly distorted my world view and lost the ability to deal with things that do not lie in the cartesian plane...unless I abstract from that plane a space, and from that space a fouth axis and so on. It is mighty hard to carry on socially meaningful conversations while maintaining this four-axis nomenclature, and when we get too far from formulations that help the average primate acheive the holy trinity of beer, broads and boob-tube we risk having frittered our time unwisely.

Chomsky's is a one stage model:

     GRAMMAR -------------------->SENTENCE

whereas I would claim that the minimum is a two stage model:

     GRAMMAR ------>SPEAKER------>SENTENCE

I think I understand possible motivation on Chmosky's part; elegance in modeling comes to mind. But sometimes the search for elegance leads to eliminating distinctions without which the problem set is unassailable.

[] static link
writebacks: 0 (writeback = trackback +/- comment)

967 - Instant Gratification

One thing that seems to drive interest in blogs is newness. Like headline news we only care about what is now. But keeping up with the latest ramblings of your favorite twenty-something cyber-curmudgeon is only the skin of the apple. Surfing yesterday's, or last year's, space can bring inspiration, can trigger the production of that next essay you have been waiting to have pulled from you like a boil you've been waiting to lance. At least that's what happened to me today when I visited Rick Crawford's vroop.com this morning. (I was actually looking for examples of ways to make my blogs look better.)

Back in the fog shrouded past of January 2004 Rick blogged a bit that could have been posted any time after 1995, a bit that will be relevant for decades to come. The overt complaint is about programmer's block, a code slinging analog of writer's block. But he bullets up a little list of specific complaints that quite put the finger on part of what is doing the blocking: Loss of focus, Attractive Distractions, Procrastination, Resentment at simple linear tasks. I see all of these as symptoms of web poisoning. There's a piece from alertbox, that I think is relevant here, although I develop the idea a little differently than the discussion in the article.

The web is addictive, much in the way a slot-machine is addictive. Pscyh research shows that the way to get conistent behavior is with inconsistent rewards. When pushing a lever always delivers food the rat pushes less often than when the lever pays off randomly. Now apply that to surfing the web or even checking your email (especially if you've subscribed to a tasty list or two.) There is a highly inconsistent reward structure; sometimes your surfing and reading will bring wild rewards of delight, emotional or intellectual jackpots. Often as not, if you were to keep track click-by-click, your surfing and reading is like dropping your nickle in the slot, pulling the arm, and watching the little wheels spin and getting nothing.

This ties back to the observations in the alertbox article. We surf for jackpots. And the medium itself, the web, with it's formatted links visually punching you in the nose with the difference between boring old data versus something that might take you to your next jackpot, reinforces this Attention Deficit Disorder addiction.

So it ain't just you, Rick, or me, it's all of us, it's the nature of our medium. It's increasingly less about Attention Deficit Disorder and increaslingly more about abuse of our thirst for valuable information by would-be purveyors of same. The web can be every bit as much a fruitless time-sink as television, but because of the one-armed-bandit effect of hyperlinks it is all the more compelling (and not nearly so socially approved...yet.)

As for what to do? Good question. I would think a regimen of hearty physical exercise, enforced "step-away-from-the-keyboard" breaks, and a routinized religious/spiritual/deep-breathing/meditative practice every 60-120 minutes would be a big step in the right direction. I pride myself on not owning a television, because I know how many of my prescious finite heartbeats I would waste every week staring vacuously at it. Will the time come when I boast similarly of not being online? Not likely, not while I'm still committed to seeing the web as an ever improving method for the sharing and stimulating of new ideas.

But there are days when I wish it was about sweeping the temple and stacking the rice bags.

[] static link
writebacks: 0 (writeback = trackback +/- comment)

968 - Well Meaning

In Linguistics, what counts as scientific evidence?

Is the epistemological choice to privilege scientism antithetical to the linguistic exploitation of intuition-reliant methods (in the sense of the 1999 Geeraerts article)? I believe the point of that article was to argue in favor of viewing intuition-reliant methods, under the rubric of hermenuetics, as complementary with methods which shun the Well of Intuition.

The confusion here lies, mostly, I think, with irregularities of English. He rocked her is not ambiguous with Throw a rock at her is what he did. But the verb in question, means, looks the same whether Fred is meaning to communicate or whether there is some etheric, phlogistonic stuff called meaning. In this way X is the meaning of what Fred said invokes this stuff in a way that Fred means X does not. At least that's how the intutions of this native speaker run. (And, thank God, an appeal to un-reason admits of no argument.)

The whirl-a-gig about "meaning" was never my intended focus; the value of treating varied approaches as complementary rather than as competing was the important part. But for those who can only skin their cat one way, that's cool too.

[] static link
writebacks: 0 (writeback = trackback +/- comment)

Home

email

Bookshelf

RSS