October 27, 2007

Ten Blinding Insights of the Japanese Newbie...

Quico says: I suppose most people think learning a new language is about as much fun as dental surgery, but I've always liked it. I learned English when I was a kid, French in High School and taught myself Italian and a smattering of Dutch after that. This year, though, I decided to seriously kick it up a notch...or, well, 9 or 10 notches. Back in January, I started teaching myself Japanese.

No, really...

This "hobby" has taken up more and more of my time and, as y'all have doubtlessly noticed, it's kind of displaced my other hobby lately. (Though, in all honesty, blogging long ago went from "hobby" to "pathological obsession" for me, so I think of the Japanese thing as detox, to a certain extent.)

Granted, I haven't even tried to learn the crazy squiggles that pass for writing in Japan, I've just concentrated on speaking and understanding. And when I say I "taught myself" mostly I mean I let the fantastic folks at JapanesePod101 teach me. (It really is a terrific website...Piitaa-san arigato!)

Still, it's been some undertaking. I struggle to think of the right metaphor for it. Lets try a couple...

Learning a language as radically heterogeneous as Japanese is a bit like putting your brain in a bread maker...it kneads it and bends it, stretches it and bakes it in ways you never imagined possible.

Or, it's like deciding to put together a 10 million piece jigsaw puzzle, an undertaking so transparently ludicrous you can't really believe you'll succeed even if you know, theoretically, that if you just keep putting together a few pieces each day it's a mathematical certainty that you'll eventually get there.

Some day. Some very, very distant day.

Over the last year, my journey's been punctuated by little epiphanies about what Japanese is and what language is and how it is and how it isn't and about what I can do and what I can't do. Thing is, my perceptions about these things keep changing. Looking back, the first few insights seem pathetically primitive now, but still, at the time they felt momentous.

So here, for I'm-not-really-sure-whose-benefit, is my take on the Ten Blinding Insights of the Japanese Newbie...

Insight #1: It's not Chinese!
(Well, I told you they start out real basic!)

This one's actually more subtle than it seems. Japanese and Chinese look the same, because the Japanese stole their aforementioned crazy squiggles from China a millennium and spare change ago. On a map, the two countries look like they're kind of near to each other, so the tendency is to figure that the languages must be related somehow, like Spanish and Italian, or at least like Spanish and English.

But it turns out that Chinese and Japanese are totally unrelated, as different from each other as each of them is from English. In fact, Japanese is a language isolate: it's related to nothing (unless you count Okinawan.)

It took me about 20 minutes of Japanese language study to reach this first insight. It dovetails directly into the next one, namely...

#2: Hey, I can actually pronounce this stuff!
Unlike Chinese, but like European languages, Japanese is not tonal. Tone does not carry meaning, so Japanese has none of the sing-song quality of Chinese or Vietnamese.

Better yet, Japanese phonetics are remarkably similar to Spanish phonetics. Aside from the vexed questions of Ls and Rs (which they don't differentiate) their sounds aren't really that different from ours. So you're hardly ever stomped by a sound you can't quite hear, and if they can say it, you can say it. That's pretty positive stuff, and it gets even better, because when you start studying you very soon realize that...

#3: This language is ridiculously simple!
If you've seen The Last Samurai, you already know this. Japanese grammar is weirdly, disarmingly simple. There's no singular and plural, no masculine and feminine, no future tense. Verbs don't conjugate. (It's all "I am, you am, he am, we am...") There are no articles. (No articles!) And no true pronouns either. Actually, most of the things that make European languages hard are utter non-issues when you're learning Japanese.

This is an epiphany that takes maybe a couple of weeks to reach. And it's great! At this stage, the task ahead seems positively easy.

The only problem is that if you keep working on it for more than two weeks or so, the very next insight is:

#4: This language is ridiculously complicated!
I was worried about verb conjugation? Ha! These little slant eyed bastards have much, much nastier tricks up their sleeves!

See, this whole mismatched-gramatical-categories thing turns out to be a two way street. Sure, they don't have some of the categories we have, but by the same token we don't have some of the categories they have. And they're weird!

Where to start? Japanese has something called copulas, which are sentence ending wordlets that indicate...well, it's not exactly clear what they indicate, but they have something to do with levels of politeness (on which a whole lot more later).

Then there's the word-ending particle, bane of all novice students. Instead of word order, Japanese relies on quirky little one-syllable suffixes to convey a word's grammatical role. If it's an object, you stick a -wo at the end, if it's a subject, you use -ga, and if it ends in -wa then it's the topic.

(Oh, did I mention? They have something called "Grammatical Topic", which corresponds to...nothing at all in European languages. More on particles later...)

Long story short, the elation you get from insight #3 dissipates pretty damn fast. You soon realize that you've not only volunteered to put together a 10 million piece jigsaw puzzle, but the pieces exist in 7-dimensional-space.

It's at around this stage that you start to develop a sense of community with every Japanese person who's ever had to learn a European Language. You start picturing them as they hit their own level 3 insight ("it's sooooo easy, Takahashi...no copulas, no grammatical topic, they don't even have suffix markers...total piece of cake!") And it's around this time, as you start to imagine what this picture looks like from the other side, that insights 3 and 4 meld together into a new, higher level insight...

#5: Everything that's hard in European languages is easy in Japanese, but everything that's easy in European languages is hard in Japanese!
This, I think, is the insight that just keeps on giving. Every single day I find fresh evidence to support it.

Maybe two months into the learning process, you start to develop an awareness that, in Japanese, the problems are never going to come from "the usual suspects." The stuff that's going to give you no end of grief is always going to be the stuff that seems most innocuous.

Take the pronoun "I". What could be simpler? What could be more basic? How, pray tell, do you say "I" in Japanese?

If you look it up in the dictionary, it'll lie to you and tell you that "I" is "watashi".

That's a half truth at best, and can be plain wrong in certain circumstances. "I" can be "boku" if you're a guy talking to other guys, "ore" if you're boasting, "atashi" if you're a teenage girl, "watakushi" if you're trying to be extra polite and two or three others that slip my mind just now. It's a plain old nightmare.

Turns out, the only honest answer to the question "how do you say 'I' in Japanese?" is "it depends". Who are you? Who are you talking to? Are you a man or a woman? Young or old? Happy or pissed off?

In Japanese, even the most rudimentary sentences are relational: they vary depending on the social situation, on the relative status of the participants, their genders, age, etc. So there are always a half-dozen ways you can translate even the simplest English sentence into Japanese. Trouble starts right from the word "I".

Of course, you totally picture how this would play into Japanese people's own Insight #3s.
"Hey, did you know there's only one word for 'watashi', 'ore', 'boku', 'atashi' and 'watakushi' in English?"

"Masaka! English must be dead easy!"
Come to think of it, if Japanese people have their own insight #3, they must also have their own #4 and #5s, no? It's a thought that solidifies your feeling of camaraderie with every Japanese person who ever struggled to learn a European language. You start to get a sense of the scale of the challenge they face, because you're climbing that same mountain, just from the opposite side. This idea yields an enormous, awed sense of respect for anyone who's ever managed to bridge the gargantuan chasm that separates these two language groups.

Anyway, the list of impossibly simple things that make your life difficult in Japanese is long. There's no ifs, ands or buts about it.


To be precise, there's no one way to say if, and or but. A word like "and" will change depending on the context. "I went to bed and turned off the light" takes a different "and" than "I like pears and bananas." "But" can take at least four different forms (demo, kedo, keredo, ga). And you could fill a book with the different ways of glossing "if".

In fact, at times it seems like the entire Japanese language is a perverse experiment in trying to dream up new ways to make incredibly simple linguistic constructs excruciatingly complicated. Don't even get me started on counting. Any language where the word for "six" changes depending on whether you're talking about six apples, six people, six buildings, six years or six elephants was clearly invented by the clinically perverse purely for the pleasure of watching foreigners get it wrong. It's just plain pointless.

(Though, to carp on the symmetry of it all once more, it's a quirk that strikes Japanese people as neither more nor less pointless than this bizarre fetish Europeans have with that mysterious linguistic entity of negligible utility known as the "article.")

At some point, you try to say something like:
"Hello, I am Francisco. I am 31 years old and I live in Europe but I was born in Venezuela"
and you realize that you can't say it, but not because you can't say "hello" or "live" or "born" but because you don't know how to say "I" or "and" or "but" or even "31".

So you whimper off, depressed, cursing every Tenno who ever lived as you become deeply, intimately convinced that you will never ever figure this language out.

This is a discouraging stage. You sit there, struggling to memorize unfamiliar words that all look alike. It's slooooow going. Then, at some point (this happened around three months into it for me) you hit the pits. The absolute low point of demoralization and just sheer discouragement. It's the dreaded, dreaded insight #6...

#6 It's just sooooo different...
"Well, duh" right? I mean, it's obvious, isn't it? You don't need to spend three months studying Japanese to know that it's totally different from any European language. Everybody knows that...

Well, yes and no.

Before you study Japanese, you certainly know that Japanese is totally different, but you can't feel it. The sheer weight of the difference, the pervasiveness of it can't really sink in until you've made a serious effort to learn a bit and found yourself foiled at every turn by the most embarrassingly minor (seeming) details.

Of course, it's a hopeless feeling to try to explain. If you haven't been through it, any explanation will be insufficient, if you have been through it, every explanation is superfluous. Meaning is built out of words in ways that are just totally alien to western ears. Nothing ever seems to fit together in the way you might expect. It doesn't matter how many times people try to explain unidirectional verbs (like "ageru", "kureru" and "morau") to you, some deep, primitive part of your lizard brain just refuses to accept that a word can function that way. Whaddayamean I can morau something from you but you can't morau something from me? It's infuriating! Hell, it's more than infuriating, it's an outrage against common sense!

I guess most people quit when they get to this stage, and the urge is perfectly understandable. But my ego bruises easily, and I couldn't accept defeat after just a few months. So I took a couple off weeks off to clear my mind. And it was during that break that I had my most platitudinous seeming (and, therefore, also deepest) insight yet...

#7: It's no use fighting it
See, when you're on insight #6, you manage to convince yourself that somehow, it's personal. That Japanese is being difficult out of spite, that it's determined to humiliate you, that it goes out of its way to prove to you that you can't learn it. This is, of course, a totally insane idea, but that's how you feel around month four.

But somehow, during my break, the sheer self defeating lunacy of the thought became clear to me. I was fighting the language! But whingeing constantly about how hard it is doesn't somehow make it easier...it only stresses you out. In a flash, I realized that every minute I spent fighting the language was a minute I wasn't spending learning it.

When this insight hits you, you have to stop. It's a time to collect your thoughts, to take the measure of the task ahead of you all over again. Truth is, if you think you can learn Japanese in a year or two, you're setting yourself up to fail. You can't! And getting upset about it won't change that reality.

If you're studying outside Japan, the reality is you need 3 or 4 or 5 years to learn passable Japanese. That's all. It's not its fault that it takes a long time to learn. It's not your fault that it'll take you such a long time to learn it. It's nothing personal. It is what it is.

All you can do is keep putting together bits of the jigsaw, keep hammering away at it and know that some day you'll get there. What's empowering here is the certainty that, so long as you persevere, you really can't fail.

When you change your time perspective like this, you finally give yourself a decent shot at success. The key is to think through where exactly the blockages are and work through them patiently and with a bloody-minded relentlessness.

And for me, at that point, definitely the biggest blockage was all about syntax. Which takes us to...

#8: Even if you understand every single word in a sentence, that doesn't necessarily mean you understand the sentence!
This insight is almost as discouraging as #6. You can sit there, listening to your lessons, and even though you know what every single word in the sentence means, the overall meaning is lost on you.

You might hear "Joe, Andy, Wife, Cousin, Chocolate Cake, New Car, Coffee Shop, Eat, Go," in that order, each with its own little word-end particle. You might even understand each word separately. But you have no idea what it means!

If you miss even one particle, you have no clue if Joe is driving Andy's car, or the cousin's wife is driving Joe to Andy's coffeeshop to eat chocolate cake, or if it's the cake that wants the coffeeshop to drive the new car and eat andy's wife with cousin Joe. Knowing all the words isn't nearly enough until you get a knack for how they hang together. Without it, you're lost.

For the same reason, you can't actually build a sentence, even if you know each of the words you want to put in it. It's infuriating! It's a hump you have to overcome...but at first it's far from clear how to.

The reason is that word order in Japanese is hideously counterintuitive. All you know for sure is that the verb goes at the end. Other than that, it's chaos out there...synctactic anarchy.

The grammar books actually make it worse by telling you that, unlike in English, it doesn't matter which order words go in...which only makes everything even more confusing. But let me give you an example so this all gets a bit less abstract.

In English, if you read a sentence like:
"Joe eats cake"
you know that "Joe" is doing the eating and the cake is doing the being eaten simply because of the order the words are in. But in Japanese they have those suffixes, which means that the words can go in any order they like. So,
"cake-wo Joe-ga eat"
"Joe-ga cake-wo eat"
are exactly equivalent, because Joe is marked by "ga" and the word that has the "ga" does the action. Of course, that also means that if you say,
"Joe-wo cake-ga eat"
the cake just ate Joe! It doesn't enter into it at all that you said "Joe" before you said "cake". Mess up your word-ending particles, and you hideously mangle the meaning of any sentence.

Of course, with a very simple sentence like Joe eats cake, it's not such a problem. But what if you want to say, oh, I dunno something like "Joe is going to the coffee shop in Jim's new car to eat chocolate cake with Andy and his wife's cousin"? Suddenly this arbitrary syntax thing becomes an absolute bloody nightmare. Trying to keep track of all the little suffixes, of who did what to whom, trying to figure out who initiates the action seems nearly impossible. At least, that is, until you hit insight #9...

#9: The subject goes first, the verb goes last, and everything in between is backwards!
Now, any grammarian reading this probably did a spit take just now. "Dear God, that's not right!" And, I admit it, it's not really right. It's a rule of thumb. A heuristic. It has a job that's more important than being right: it makes your life bearable.

The grammar books probably are right in that, in some formal sense, word order in Japanese really is free. But when you're learning, you don't need academic precision, you need a lifeline to rescue you from drowning in the sheer anarchy that is the Japanese sentence's syntax. And this rule of thumb gets you safely to syntactic shore 95% of the time.

Take the nightmare of a sentence above,
"Joe is going to the coffee shop in Jim's new car to eat chocolate cake together with Andy and his wife's cousin"
Where to start? Well, first you find the subject (in this case "Joe") and you put it first. Then you pick out the main verb (in this case, "is going") and stick it at the end. Already you know the sentence is going to look something like:
"Joe-ga ... is going."
And how do you fill in that dot-dot-dot? Easy! You work your way backwards, clause by clause, through the English. What do you get?
"Joe-ga, Andy and his wife's cousin together with, chocolate cake to eat in order to, Jim's new car in, the coffee shop to, is going."

Bingo! A perfectly formed Japanese sentence! Study Japanese long enough and sentences like that start to make perfect sense to you.

Insight #9 gets you from knowing a bunch of words that you have no clue how to string together to knowing a bunch of words that you can kind of use.

I'm sure in the years ahead I'll understand all the nuance this little rule of thumb misses. When that happens, I'll discard it. But, for now, I think I'll keep my lifeline thank you very much. I'm not much of a nihongo swimmer quite yet, and I'm pretty sure I'd drown without it.

Of course, just as you're feeling pretty good about making your first few sentences, Japanese finds a new way to baffle you. The most prominent of these has to do with politeness. It's a huge stereotype, of course, the whole obsessive Japanese politeness thing. But what a lot of people don't realize is how finely structured the politeness-level system is, how deeply embedded in the language.

Around month five, I started to become aware that there is always more than one way to say any given thing in Japanese. So I asked my Japanese friends just exactly how many levels of politeness there are in Japanese. Seems like a reasonable question, doesn't it? Right away, though, we came up against one of those little impasses of mutual misunderstanding that serve as surefire signposts to deep cultural chasms. "What do you mean 'how many levels are there'?"

"Well, you know, in Spanish and French there are two. Tu and usted. Tu and vous. In English there used to be two, but 'thee' fell out of usage so now there's just you. I guess in Japanese there are more, but how many, exactly? Three? Four? More?"

This only got me more puzzled looks until I realized I'd have to go back and think this one through from a totally different vantage point. Eventually, just in the last few weeks, an insight crystalized. There isn't a straightforward answer to my question, because...

#10: European politeness is binary, but Japanese politeness lies along a continuum
In European languages, it's fairly straightforward. You're either using the 'Tu' form or you're using the 'Usted' form. The only thing that changes between them is verb conjugation. If you mix the two, you're just not speaking properly.

But Japanese politeness is much more multifaceted. You have the basic -masu verb form (for polite speech) and the dictionary verb form (for casual speech) but the story hardly ends there. Verbs change, but nouns and prefixes and suffixes change as well. As you go from more to less polite, more and more words and bits of word "fall away". What's worse, men speak differently from women.

The hardest thing to convey, though, is that it isn't just that there are different levels of politeness, it's also that there are also different flavors of politeness. The polite language that's appropriate in business is not the same type of polite language that's appropriate when talking to a very old person, or to a priest. The honorific language you use when (or well, realistically, if) you ever meet the Emperor is something else altogether. And the casual language you speak to women is different from the casual language you speak to men, which is different from the casual language you speak to children, which again is different from the casual language you speak if you're a criminal, or a dirty old man, or if you're very angry.

Each register has its own words, its own phrase construction patterns, its own conventions. The possibility to mix and match between them is basically unlimited. So, in a way, Japanese people are constantly, unconsciously, automatically "fine tuning" the level and flavor of politeness they're using to fit the situation they're in, to project the type of persona they want to project, to situate themselves in their social environment.

This, I think, is the reason Japanese people have the feeling that foreigners can never truly learn their language. The language is embedded in the social system to a degree people in the West can hardly conceive of. So it's not that you can't learn the language. You can. It's that you can't truly master the judgment to know exactly where on the politeness spectrum it's appropriate to speak, even if you know all the words. Even if you manage to speak perfectly grammatically, you almost certainly won't ever sound fully natural.

Like old gringos in Venezuela who continue to screw up their masculine and feminine nouns after decades in the country, foreigners in Japan can go for decades without quite grasping that sixth sense Japanese people learn as kids and allows them to know which register they need to use when.

This multiplicity of registers also accounts for one of the weirder things about Japanese. This one I'm still struggling with, but as far as I can figure, there aren't any swear words. Or, well, not as we think of them. They have a taboo register, but not specific taboo words. If you go for the very, very rudest register of them all, everything you say comes out as swearing...or, everything you say has an impact on the listener similar to the impact swearing has in the west. But it isn't any specific word that's making the speech "sweary", it's the whole package that's out of bounds. Only hardened criminals and yakuza types talk like that.

Anyway, that's the way it seems to me at this point. But then, this one hasn't quite crystallized into a proper insight yet, and if you've read this far, you've certainly realized that a big part of the way this game works is that later insights often modify or even overturn earlier ones. That's learning, baby, ask Maria Montessori if you don't believe me! Probably a year from now this write up will look totally idiotic to me.

It's been a trip learning Japanese. But I still have a long, long way to go. I'm working mighty hard to transition from Advanced Beginner to Intermediate level right now. I still make a lot of mistakes, and I have the impression I sound kind of like Tarzan when I talk. I know full well that it'll take another three or four years before I can speak presentably, but then, that is a reality I'm through fighting.

If you're thinking of learning Japanese, or any radically different language, all I can say is: slow and steady wins the race. I'm convinced anyone can learn any language, given a net connection and enough perseverance. The reason people fail is that they set unrealistic goals and get discouraged when results are too slow in coming. I blame deceptive marketing. Anyone who tries to sell you a book by telling you you learning a very different language is "easy" or "quick" is just plain lying.

It's not easy. It's not quick. But it's also not impossible.

Just remember: if you're realistic about your time horizon and grimly determined to study every single day, you can't fail. Or, well, that's what I tell myself every day, anyway.

October 23, 2007

Philip Converse: The nature of belief systems in mass publics

Quico says: It's been hard to find the time or the stomach to blog recently. (Somebody I know might have taken up the slack, but hasn't...) Rather than leave the blog dormant, I thought I might as well post some of the more provocative stuff I've been reading. Today, it's a summary of "an oldie but a goodie": Philip Converse's groundbreaking 1964 study of voter cluelessness. Even if you've never heard of it, it's a paper whose influence has permeated our political culture. It launched a sub-field in political science, and lives on with silent strength in the conduct of almost every election campaign of the last 40 years. For a gripping discussion on the implications of Converse's paper, check this out.

Wikisum (summing up Converse) says:

A great majority of people neither adhere to a coherent and complete set of beliefs able to sustain a clear ideology nor do they have a clear grasp of what ideology is. This is measured by a lack of coherence in responses to open-ended questions. What's more, the ideology of elites is not mirrored by the mass public.

Converse analyzes open-ended interview questions to measure conceptualization of ideology. He concludes that the liberal-conservative continuum is a high level abstraction not typically used by the man in the street because of response instability and lack of connections made between answers. There is no underlying belief structure for most people, just a bunch of random opinions. Even on highly controversial, well-publicized issues, large portions of the electorate do not have coherent opinions. In fact, many simply answer survey questions randomly.

Though some political sophisticates do structure their opinions in a larger ideological framework, such structure is rare. This level of political sophistication (one's "level of conceptualization") is correlated positively with the respondent's level of education, degree of political involvement, and amount of political information, but crucially also dogmatism and political inflexibility.

Democratic theory assumes that voters in the "mass public" hold clear ideological values which allow them to make voting decisions based on the positions candidates hold. One of the most prevalent distinctions they are assumed to make is evaluating candidates' positions on the liberal-conservative ideological spectrum. Thus, when the electorate chooses politicians that vary from one end of the spectrum to the other, it is often assumed that the electorate is becoming more conservative or more liberal.

The result of Converse's surveys and analysis cast doubt on many of these assumptions by showing the apparent lack of understanding of ideology or even differentiation between the two political parties on the liberal-conservative continuum. Using open-ended interviews as well as survey data, Converse classifies voters into the following categories based on their understanding of basic ideological differentiation between ideas:
  1. Ideologues: These respondents relied on "a relatively abstract and far reaching conceptual dimension as a yardstick against which political objects and their shifting political significance over time were evaluated" (p.216).
  2. Near Ideologues: These respondents mentioned the liberal-conservative dimension peripherally, but did not appear to place much emphasis on it, or used it in a way that led the researchers to question their understanding of the issues.
  3. Group Interest: This group did not demonstrate an understanding of the ideological spectrum, but made choices based on which groups they saw the parties representing (e.g. Democrats supporting blacks, Republicans supporting big business or the rich). These people tended to not understand issues that did not clearly benefit the groups they referred to.
  4. Nature of the Times: The members of this group exhibited no understanding of the ideological differences between parties, but made their decisions on the "nature of the times." Thus, they did not like Republicans because of the Depression, or they didn't like the Democrats because of the Korean war.
  5. No issue content: This group included the respondents whose evaluation of the political scene had "no shred of policy significance whatever" (p. 217). These people included respondents who identified a party affiliation, but had no idea what the party stood for, as well as people who based their decisions on personal qualities of candidates.

Most people fall into the lower three levels of conceptualization.

Converse also found that the mass public does not seem to share beliefs in any predictable way with elites. He doubts that the voting patterns of the people at the lower end of the scale follow the patterns of the ideologues and near ideologues who have a firm grasp of the issues.

In addition, Converse's interviews with the same respondents over a two-year period often show little correlation with each other. In these cases, only 13 out of 20 managed to locate themselves on the same side of a given controversy in successive interviews. Converse's interpretationis that this change seemed almost exclusively random instead of as a response to changing beliefs.

Discussing some of the implications of Coverse's research, Jeffrey Friedman writes:
Converse found that only about 2.5% of the public (as of 1956) was passably knowledgeable about the meaning of liberalism and conservatism, the “belief systems” that structured, and still structure, most political debate and public-policy making. That would be bad enough; but surely knowing what the dominant belief systems “mean” isn’t sufficient to make well-informed political decisions.

Consider the most reviled pundit on the other side of the political spectrum from yourself. To liberal ears, a Rush Limbaugh or a Sean Hannity, while well informed about which policies are advocated by conservatives and liberals, will seem appallingly ignorant of the arguments and evidence for liberal positions. The same goes in reverse for a Frank Rich or a Paul Krugman, whose knowledge of the “basics” of liberalism and conservatism will seem, in the eyes of a conservative, to be matched by grave misunderstandings of the rationales for conservative policies.If Limbaugh, Rich,et al., turn out to exemplify the “cognitive elite,”we are in serious trouble. Converse, I believe, showed just that.

Converse’s political elites are particularly well informed about what it means to be a conservative or a liberal, and their reasoning about politics is structured by this knowledge. But Converse’s findings suggest, I think, that their relatively high levels of ideological knowledge are due to their being conservative or liberal ideologues: closed-minded partisans of one point of view. Should the leadership of public opinion by such people be a source of relief—or a cause for anxiety?

Ideological constraint is a form of determination. Converse equated it with “the success we would have in predicting, given initial knowledge that an individual holds a specified attitude, that he holds certain further ideas and attitudes.” There would be nothing worrisome about such determination if people’s political attitudes were being constrained by logic or evidence. But Converse made it abundantly clear that that is not the type of constraint he had in mind.

“Whatever may be learned through the use of strict logic as a type of constraint,” Converse wrote, it seems obvious that few belief systems of any range at all depend for their constraint upon logic.” Ideologies are only “apparently logical wholes,” and the appearance is skin deep.

For Converse "what is important is that the elites familiar with the total shapes of these belief systems have experienced them as logically constrained clusters of ideas.” But this experience does not stem from the ideologue’s astute reasoning or her keen investigation of reality. Her views are, instead, determined by the political belief system she has been taught.
And later on:
Converse’s ideologues form a cultural elite, not necessarily a “power elite.” The members of the public who are sophisticated about politics (relative to most people) aren’t necessarily those who are in charge of the government. Realistically, however, the cultural elite, through its teaching and journalism, is likely to shape the ideas of the governing elite, who tend to be a highly educated subset of the cultural elite. The power of Converse’s politically sophisticated stratum, then, lies not only in its attenuated trickle-down influence on mass culture, but in its subsumption of, and influence upon, those who directly shape public policy through their positions in the legislature, the judiciary, and the bureaucracy.

In a democracy, all “branches”of government are nominally subordinate to the people. The power of public opinion is supposed to check even the nomination and confirmation of judges and the appointment of top bureaucrats, since that is done by politicians beholden to the electorate. This nominal barrier to elite rule seems to have stymied the movement to “bring the state back in” to political analysis. During the� 80���s, the “state theory” movement was poised to take political science by storm while it produced penetrating analyses of pre-democratic states, both premodern and newly industrialized, modern democracies seemed to stop “state theory” in its tracks.

How,after all, can state personnel act autonomously if, as in modern democracies, “society,” or public opinion, controls the flow of revenue that pays for the standing armies that, in classic state theory, undergird “strong states”? The key role played by the military (and its supportive tax-collecting bureaucracy) in state theory—putting down popular protest—fades to insignificance in democracies, where public disaffec-
tion is translated into change nonviolently,through the ballot box—and where the ballot box controls the military. Is state autonomy possible, then, once the state is democratic?

Post-Converse public-opinion research can provide a positive answer to this question: The public can’t control what it doesn’t know about. But since that research is usually the province of specialists in American politics, this answer was not apparent to state theorists, who tended to be comparativists. (And the question did not seem to occur to public-opinion researchers, for the reverse reason.) Only recently has the cross-fertilization of public-opinion research and state theory begun, based on the simple premise that a public as ignorant as the one portrayed by Converse is unlikely to be aware of most of the things its government does. Public ignorance maythus sever a democratic state from the demos.

Government officials can let their own ideological agendas shape bureaucratic rule making, judicial decision making, and the crafting of legislation, without fear of electoral reprisal—even if their agendas are unpopular—to the extent that they think that the public is unlikely to find out about it. State autonomy in democracies, then, would have to do not with the efficiency of tax collection or the reliability of armies,
but with a government so big that nobody can keep track of its activities.

This separation between people and policy has limits. The mass media can, by pounding away at a proposed or extant government policy, bring the public’s limited attention to it. Hence Bush’s defeat on Social Security. But these limits have limits: very few issues will receive such sustained media attention that they provoke public outrage and, thus, negative political consequences for unpopular policies.
Friedman's a terrific writer. It's worth reading his entire spiel.