ࡱ> 7 SkbjbjUU &7|7|Sgl Lbbbbbbbh j j j j j j $ R bbbbb bbb bbbbbbh bbh bb\ \ b  MA(b\ \ 0 \ b\ b The great IQ-score rise: test sophistication, Caesarian births or masked mystificatory multipliers? For the past twenty years, the New Zealand political scientist James Flynn (1980, 1987, 1992, 1995) has been trying to show that the large (15 IQ-point) Black-White difference in intelligence need not have a genetic explanation. Re-discovering and confirming the twentieth-century rise in IQ-test scores that had first been found by Cattell (1949) and Koppen-Thulesius & Teichmann (1972), Flynn has noted there must be substantial environmental influences on IQ scores even if these secular influences, which he sums up as 'X'-factors, do little to explain differences between individuals or between groups which belong to the same generation. Thus Flynn has held out the hope that the final discovery of X-factors might allow a rapid boost to be given to Black IQ. Recently, Flynn (Dickens & Flynn 2001a, 2001b) has become more confident of the existence of X-factors for he has noticed what psychogeneticists have routinely maintained for thirty years (e.g. Brand, 1996/2001, Chapter 3), that genetic influences on IQ have their effect partly by causing an individual child to have a special and perhaps supportive micro-environment as when bright children choose to read a lot and are then rewarded and encouraged to undertake still more reading. Flynn thus maintains that features like libraries and violins are "masked" environmental influences; and he thinks that "social multipliers" might allow such influences to have really big effects perhaps sufficient to raise Black IQ if only Black culture and society could be changed. Thus Flynn thinks he can imagine new environmental influences that could fill the gap left by the genetic influences championed by psychologists like Hans Eysenck and Arthus Jensen and the macro-environmental, cultural and between-family influences preferred by social scientists like Franz Boas and Ruth Benedict. There are six main problems with Flynn's argument. 1. The secular IQ-score gain has long been argued to be the result of increasing test sophistication (e.g.  HYPERLINK "http://www.douance.org/qi/brandtgf.htm" Brand 1996/2001, Chapter 4). Certainly, the gains on some IQ subtests are greater than the gains on others: they are greatest on tests which place a premium on speed rather than accuracy and otherwise on those tests that are less g-loaded (Rushton, 1999). 2. Flynn himself has often written that IQ scores do not measure intelligence but only "a weak correlate" of intelligence, possibly "mere problem solving ability." In his 2001 position, he abandons such thinking without any explanation. For many years, he maintained that intelligence was not truly rising because university professors noticed if anything the opposite and because patent applications were not increasing; yet now he enthusiastically but inexplicably hopes his secular gains hold out to Blacks the hope of improvements that will be well beyond the trivial. 3. Other psychologists (Lynn et al., 1987) have argued that part of the secular IQ rise is real but is simply attributable to improved nutrition, obstetric care and stimulation (e.g. by the availability of radio and TV in every home). The point here is twofold: that no magic multiplication is involved; and, since Blacks have already experienced such improvements (for them, relatively greater improvements from their environmental baseline than for Whites), there is no reason to believe that 'masked' environmental X-factors have been retarding Black intelligence. Recently, claims that the Black-White gaps in IQ and literacy are actually closing have been rejected by Murray (1999). 4. One can welcome Flynn's agreement that environmental effects on intelligence often occur principally in interaction with genetic effects (as when the gift of a violin delivers a result only in a child who is already musically gifted and willing to practise). But such influences cannot of their very nature be expected to deliver results in a wide range of people let alone throughout the Black population. What is necessary to make use of interaction or covariational effects is to identify the different educational moves and incentives that will be necessary for high-IQ children on the one hand and for low-IQ children on the other. The fact that increased rewards for literacy may have helped raise intelligence, especially in high-IQ children, does not mean that similar literacy incentives will work for low-IQ children. 5. Flynn's magic multipliers are envisaged to give real final potency to initially enhanced intellectual performance. Yet some intellectually challenging and once-rewarded activities, the writing of Latin and Greek verse and the playing of whist, largely died out in the twentieth-century despite plenty of encouragement from social institutions (schools and women's guilds). Moreover, rather than IQ-score gains occurring especially in the higher-IQ range, the tendency has been to find greater gains among the low-IQ (Teasdale & Owen, 1989). 6. Flynn's basic model for a utopia of social multiplication is that of sports and music: here the provision of tools and encouragement and group membership has obvious effects for a child who is suitably gifted. But such developments will necessarily occur at the expense of the child's performance in other areas, since time for practise is finite. Few Olympians are all-rounders. However, this model is quite unsuitable for understanding general intelligence or any real gains in it. For, notoriously and remarkably, general intelligence predicts success in diverse mental abilities: there has never been any reason to believe that there are influences which increase some mental abilities only at the cost of impairing others. Altogether, Flynn's initiative is very welcome in drawing attention to: (a) substantial genetic influence on IQ and (b) relatively slight influence of macro-environmental features on IQ. Moreover, (c) the scholarly and optimistic tone that Flynn has maintained is a pleasure in this controversial and much-censored field; and (d) he has alerted psychologists to a secular rise in IQ scores which they themselves had tended to neglect mistakenly maintaining for years that recidivist criminals had normal IQs rather than the mean IQ of 85 that is recognized today now that updated test norms are available (Gordon, 1986). However, Flynn's attempt to escape from admitting a largely genetic basis for the Black-White difference must be judged a failure; and his implication that 'high-IQ-environments' should be supplied to low-IQ children in a Head Start Mark II lasting virtually for life is a positively dangerous illusion which can only risk further resentment of wasteful expenditures to provide salaries for egalitarian educationalists. Fortunately, an alternative but realistic optimism is possible: experts have been recognizing that children of all ability levels can be helped by tailoring teaching to their individual IQ levels and allowing them to proceed through school at their own speeds ( HYPERLINK "http://www.douance.org/qi/brandtgf.htm" Brand, 1996/2001; Benbow & Lubinski, 1996; Brand, 1998). References BENBOW, Camilla P. & LUBINSKI, D. (1996). Intellectual Talent: Psychometric and Social Issues. Baltimore and London : The Johns Hopkins University Press. BRAND, C. R. ( HYPERLINK "http://www.douance.org/qi/brandtgf.htm" 1996/2001). THE g FACTOR. Chichester, UK : Wiley DePublisher. BRAND, C. R. (1998). Fast track learning comes of age. Personality & Individual Differences 24, 6, 899-900. DICKENS, W. T. & FLYNN, J. R. (2001a). 'Heritability Estimates vs. Large Environmental Effects: The IQ Paradox Resolved'. Psychological Review. DICKENS, W. T. & FLYNN, J. R. (2001b). 'Great Leap Forward'. Opinion Essay in New Scientist, No 2287, 44-47, April 21st. CATTELL, R.B. (1949). The fate of national intelligence: test of a thirteen- year prediction. Eugenics Review 42, 3. FLYNN, J.R. (1980). Race, IQ and Jensen. London : Routledge & Kegan Paul. FLYNN, J.R. (1987). Massive IQ gains in 14 nations: What IQ tests really measure. Psychological Bulletin 101, 171-191. FLYNN, J.R. (1992). Achievement Beyond IQ. Hillsdale, NJ : Erlbaum. FLYNN, J.R. (1995). What environmental factors affect intelligence? - The relevance of IQ gains over time. In D.K.Detterman, Current Topics in Human Intelligence 5. Norwood, NJ : Ablex. GORDON, R. (1986). Review of J.Q.Wilson & R.J.Herrnstein, Crime and Human Nature. Business Horizons, vii/viii. KOPPEN-THULESIUS, K. & TEICHMANN, H. (1972). Accelerative trends in intellectual development. British Journal of Social & Clinical Psychology 11, 3, 284-294. LYNN, R., HAMPSON, S.L. & MULLINEUX, J. C. (1987). 'A long-term increase in the fluid intelligence of English children.' Nature 328, 797. MURRAY, C. (1999).  HYPERLINK "http://www.lrainc.com/swtaboo/taboos/cmurraybga0799.pdf" http://www.lrainc.com/swtaboo/taboos/cmurraybga0799.pdf. RUSHTON, J. P. (1999) 'Secular gains in IQ not related to the g factor and inbreeding depression unlike Black-White differences: A reply to Flynn.' Personality & Individual Differences 26, 381-389. TEASDALE, T. W. & OWEN, D. R. (1989). 'Continuing secular increases in intelligence and a stable prevalence of high intelligence levels.' Intelligence 13, 255-262. ------------------------------------------------------------------------------------------------------ REJOINDER BY JIM FLYNN TO CHRIS BRAND ON DICKENS/FLYNN (1 v 2001) A PREFACE The liberal reaction to the Bell Curve was sad in that they became so distracted by race that they did not confront what was really central: the meritocracy thesis which was quite independent of the race and IQ debate. Let me try to prevent critics from making the same mistake about our model (Let me also say: Bill deserves credit for the model and its maths. He seems to feel I made an important contribution in isolating its essentials, critical comment, and expressing its merits clearly. Perhaps). Imagine that we found the genes that determine intelligence and can show that blacks have worse ones. That would not undermine the value of the model one whit. It would merely show that the black/white IQ gap is not caused by the kind of exogenous factors that we posit (very tentatively) as causes of the generational IQ gap. We believe that the model would still provide an explanation to the paradox that has been driving us (and lots of others) crazy: how can environment be potent enough to cause massive IQ gains when kinship studies give such high h2 estimates - WITHOUT resorting the the absurd notion of a factor X. Recall that a factor X is not simply an unknown. To elucidate, high h2 estimates seem to show: Either (1) that environmental factors operating within the generations are so feeble that if they were operative between generations, you would have to posit about a 3 SD envir gap between the generations - which means the last generation had an average envir worse than 99% of the present generation - implausible to say the least; Or (2) that a factor operates between the generations that is uniform in its effects within each generation (and therefore does not show up in the h2 estimates). That is a factor X and it seems totally absurd (what could have such uniformity of effect?). Contrary to what Chris says, Flynn has never wavered in condemning the positing of a factor X as absurd. What Flynn did say, after he detailed massive IQ gains over time, was this: either factor Xs do exist - which can hardly be true; or there is something the matter with the implications we are drawing from h2 estimates - but I can't for the life of me see what. Bill found the answer. The model shows that push h2 estimates as high as you like, you can still show how environmental factors could produce huge between-group IQ differences - without the absurd uniform impact of factors Xs. And without raising IQ variance over time (another baffling problem). That should whet everyone's appetite to consult the original article - whose analysis on this point Chris in no way confronts or disputes (just as so few have confronted the meritocracy thesis - I have). The model has other virtues. As Chris says, it incorporates stuff that is old hat in the literature on the 'nurture' side - but whose implications were never seen in the above context. That is why we think it has the potential to unite me and Chris - if he will only take off the race spectacles. It also integrates things in the literature that seemed discrete and odd - h2 rising with age - intevention effects - adopted kids whose IQs eventually correlated with the natural parents' IQs - oddities arising out of Jensen's method of correlated vectors. Now my comments on extracts from Chris's comments: CHRIS >Recently, Flynn (Dickens & Flynn 2001a, 2001b) has become more confident of >the existence of X-factors for he has noticed what psychogeneticists have >routinely maintained for thirty years (e.g. Brand, 1996/2001, Chapter 3), >that genetic influences on IQ have their effect partly by causing an >individual child to have a special and perhaps supportive >micro-environment - as when bright children choose to read a lot and are >then rewarded and encouraged to undertake still more reading. ME Not true that either Flynn or the article posit X-factors. Glad that Chris believes what feeds into the model is old hat - presume that means it will not be disputed CHRIS Flynn thus maintains that features like libraries and violins are "masked" >environmental influences ME Correct - where is the reason for disagreeing CHRIS and he thinks that "social multipliers" might allow such influences to have really big effects ME Correct - where are the reasons for disagreeing CHRIS perhaps sufficient to raise Black IQ if only Black culture and society could be changed ME Corrrect, but let us debate that at another time CHRIS The secular IQ-score gain has long been argued to be the result of >increasing test sophistication (e.g. Brand 1996/2001, Chapter 4). Certainly, >the gains on some IQ subtests are greater than the gains on others: they are >greatest on tests which place a premium on speed rather than accuracy and >otherwise on those tests that are less g-loaded (Rushton, 1999). ME I think that the crucial last sentence is false, but I don't want to go over all of that again. Instead, let us assume that Chris's hypothesis is true. It is not a hypothesis about 'test sophistication' in the usual sense. It says that (using our model's language) certain environmental triggers were at work between the generations - corrupting influences like rock music and progressive education that promoted social decay and irresponsibility. These caused the present generation to be more prone to do something he calls 'intelligent guessing' which hugely boosts their IQ test performance. Perhaps Chris will object to my criptic description of the trait in question, but the precise description is not relevant to the point I am about to make. Let us call the trait 'T'. Does he want to say that the triggers of T (susceptibility to rock music, exposure to progressive education) operate like a factor X? Do they operate only between generations, or do they vary significantly with generations as well. If the former, he has embraced the mysterious factor X and not me - and must face up to the usual accusations of implausibility (certainly the triggers of decay cannot have affected no one a generation ago and everyone now). If the latter, they should show up as a powerful environmental factor in h2 estimates - but they do not - no envir factor shows up in h2 estimates powerful enough to explain the sheer magnitude of IQ gains. Using the conventional model, the present generation would have to have an average environment so 'corrupt' as to be worse than the environs of 99% of the previous generation. Is that REALLY an open option? Therefore, Chris is caught in the same paradox as everyone else who tries to reconcile high h2 estimates with powerful between-generation envir factors - and he needs something like our model as a solution. The model itself is agnostic as to specific causes - Chris's can be plugged in just like anyone else's. Just so long as a rise in the mean of the trait in question would have reciprocal causation with environment. Why not? When his triggers push up the societal mean for T (say irresponsible risk-taking - intelligent guessing?), the social multiplier kicks in. The heightened level of T will affect the socialization of everybody (more risk-taking expected in the family, work, playing bridge) - and as individuals respond, raise the mean further. So you would get a big boost in T (and in IQ test performance) in a relatively short time. No doubt reciprocal causation between T and envir is more plausible for some traits than for others. If it be thought that it would be minimal for Chris's T, our model will simply not help him get out of the fatal paradox - and he will have to get himself out using some alternative. But the paradox will not go away. CHRIS Flynn himself has often written that IQ scores do not measure >intelligence but only "a weak correlate" of intelligence, possibly "mere >problem solving ability." In his 2001 position, he abandons such thinking >without any explanation. ME Here I repeat a reply made to a jounalist who asked for an explanation. Mr Paulus - I speak for myself in that Bill Dickens has never gone on record on this point. Three things: (1) The model in our recent article presumes that IQ gains signal an increase in cognitve skills that have real-world effects. Otherwise there could be no reciprocal causation between rising talent and enriching environment such as we assume in the 'social mutiplier'. (2) I think that these cognitive gains are real and have social significance. It is certainly significant if people have uprgraded the cognitve content of everyday conversation, leisure, are better at chess, provide a larger pool of people who feel more comfortable with abstractions (perhaps potential mathematicans and physicists?), and may do more lateral thinking at work. I think it is fair to say that I have come to accept a longer list of significant real-world effects over the years. Though I never denied that the IQ gains signalled that something was happening in the real word - people who do better on IQ tests have to be better at something - they are not struck by lightening in the test room. (3) However, I am still convinced that to identify the IQ gains with INTELIGENCE gains in the common meaning of the word would lead to expectations that are false. By 'more intelligent', I mean minds that, all else (attitudes, motivation, etc.) being equal, are either quicker at learning, or learn better in the same amount of time, in areas NEW to the person. I believe two things. (a) If anyone expects schoolchildren to grasp new material as if most of them were gifted, they will be disappointed. It is interesting that the first empirical study of teachers of many years experience confirms this. (b) If anyone expects mental retardation to virtually disappear, they too will be disappointed. Yet, these things would follow from a simple identification of IQ gains with intelligence gains. To borrow the basketball analogy from our recent article, you can have a real-word, and socially significant, increase in basketball performance without a dramatic escallation of athletic aptitude - without encountering a new generation of 6-year-olds who just pick up most sports with surprizing speed! CHRIS Other psychologists (Lynn et al., 1987) have argued that part of the >secular IQ rise is real but is simply attributable to improved nutrition, >obstetric care and stimulation (e.g. by the availability of radio and TV in >every home). ME Even Lynn accepts real gains of as much as one SD and therefore, needs something like our model to sove the paradox it addresses. See comments on the 'Brand hyposthesis' made above. CHRIS One can welcome Flynn's agreement that environmental effects on >intelligence often occur principally in interaction with genetic effects (as >when the gift of a violin delivers a result only in a child who is already >musically gifted and willing to practise). But such influences cannot of >their very nature be expected to deliver results in a wide range of people ME To avoid a possible confusion, our posited reciprocal causation beteen talent and environment is not the same as what Chris may be getting at above (gens X environment interaction). But we agree to the extent that gifted musical genes may capitalize on a violin, just so long as it is clear that ungifted genes may be correlated with skipping piano lessons. Every one is affected by reciprocal causation. But, there is an important point here. As the model says, when genetic differences between individuals within a generation seize control of reciprocal causation between IQ and envir, they do NOT raise the mean performance - merely spread people out above and below the mean. It is WHEN persistent social forces operating between generations (spin-offs of the industrial revolution) seize control that they raise the societal mean IQ. CHRIS Some intellectually challenging and once-rewarded activities, the writing of Latin and Greek verse and the playing of whist, largely died out in the twentieth-century despite plenty of encouragement from social institutions (schools and women's guilds). ME Yes but so what? Let someone else explain why Greek has declined. IQ has risen and we are trying to explain that. Chris himself assumes that some kind of social trend must be boosting mean IQ. CHRIS Moreover, rather than IQ-score gains occurring especially >in the higher-IQ range, the tendency has been to find greater gains among >the low-IQ (Teasdale & Owen, 1989) ME How long will this myth persist? T&O are right about Scandinavia - many data sets elsewhere (US, Holland, Israel, Argentina) show huge gains at all IQ levels. Does ahyone really believe that the Dutch gain of 20 points between 1952 and 1982 was a matter of 60 points over the bottom third of the curve and nil elsewhere? CHRIS Flynn's basic model for a utopia of social multiplication is that of >sports and music: here the provision of tools and encouragement and group >membership has obvious effects for a child who is suitably gifted. But such >developments will necessarily occur at the expense of the child's >performance in other areas, since time for practise is finite. Few Olympians >are all-rounders. However, this model is quite unsuitable for understanding >general intelligence or any real gains in it. For, notoriously and >remarkably, general intelligence predicts success in diverse mental >abilities: there has never been any reason to believe that there are >influences which increase some mental abilities only at the cost of >impairing others. ME The analogy with basketball is just that. We never mention music or violins, by the way. Either the model makes sense of the data peculiar to the IQ area or it doesn't. Look at it and judge for yourself. In passing, note that IQ gains since 1950 have been very uneven on WISC subtests. For example, almost nil on arithmetic and vocabulary, huge on Similarities and Block Design. This is something to be explained, as I have attempted to do, not merely swept under the carpet on the altar of g. CHRIS Flynn's initiative is very welcome in drawing attention to: (a) >substantial genetic influence on IQ and (b) relatively slight influence of >macro-environmental features on IQ. Moreover, (c) the scholarly and >optimistic tone that Flynn has maintained is a pleasure in this >controversial and much-censored field; and (d) he has alerted psychologists >to a secular rise in IQ scores which they themselves had tended to neglect - >mistakenly maintaining for years that recidivist criminals had normal IQs >rather than the mean IQ of 85 that is recognized today now that updated test >norms are available (Gordon, 1986). ME Thanks - you and I rise above ideology and treat each other as human CHRIS However, Flynn's ... implication that 'high-IQ-environments' should be >supplied to low-IQ children in a Head Start Mark II lasting virtually for >life is a positively dangerous illusion which can only risk further >resentment of wasteful expenditures to provide salaries for egalitarian >educationalists. Fortunately, an alternative but realistic optimism is >possible: experts have been recognizing that children of all ability levels >can be helped by tailoring teaching to their individual IQ levels and >allowing them to proceed through school at their own speeds (Brand, 1996; >Benbow & Lubinski, 1996; Brand, 1998). ME Note that life-time programmes that would compromise a person's autonomy in favour of surrounding them with an intervention environment is an 'implication' - which is to say it is nothing that Dickens and Flynn either mention or would reccommend, but something wicked people might do thanks to reading us. I say to them, don't even think about it. Read what we actually say: that interventions (including adoptions) have short-range effects and unless we can learn how to get their graduates to target cognitively stimulating environments (and neither of us know how to do that at present), interventions must be valued primarily for whatever non-cognitive gains they engender. SUMMARY PRAYER What strikes me about Chris's comments is that (in my opinion) nothing central to the model is disputed - and therefore, there is no challenge to its ability to do the job (solve the paradox) it was designed for. Bill's model takes what is generally conceded by all and works a bit of magic. That hardly makes it pedestrian. We still hope for some steps toward unity from all persons (hes, shes, its, and theys) of good will and open mind. Those so inclined, try this experiment: Imagine that the model was published by Arthur and the late Hans Eysenck - and accompanied by comment on the limitations of intervention pogrammes - and by evidence that no systematic envir differences separate black and white, none analogous to those that separate the generations. Then actually assess the MODEL ITSELF. ghijJ K  h i hq    < G s &!'!(!1!2!5!A!!!P"d"""""?#Q#j#}##$/$D$$%W%n%p%%%%%&&&&&'''j=6U]j6U]H*6j~U\6\jU6]0JjU jUCJ Ighiij s: ; < G H I g!!f" 0^`0$a$Skf""V##$_$%%-&&L'(((( )!)")QkRkSk 0^`0 0^`0'I'J'L''(((")e)f)g)QkRkSkOJQJCJ 6]j6U] 0J6]$&P 1h. A!"#$%DyK yK Nhttp://www.douance.org/qi/brandtgf.htmDyK yK Nhttp://www.douance.org/qi/brandtgf.htmDyK yK Nhttp://www.douance.org/qi/brandtgf.htmUDyK 8http://www.lrainc.com/swtaboo/taboos/cmurraybga0799.pdfyK phttp://www.lrainc.com/swtaboo/taboos/cmurraybga0799.pdf i8@8 NormalCJ_HaJmH sH tH 0@0 Heading 1$@&CJ<A@< Default Paragraph Font@/@@ List"h5$7$8$9DH$^h`aJ>B@> Body Textx5$7$8$9DH$aJ.U@. Hyperlink >*B*ph4P`"4 Body Text 2$a$CJ Sg ghiij s:;<GHIgfV _ !!-""L#$$$$ %!%"%QgRgUg00000000000000000000000000000000000000000000000'SkX\f"SkY[SkZJ'1"#I#SgXXXX Chris Brand(C:\My Documents\07\Dickens&Flynn2001.doc Chris BranddC:\WINDOWS\Profiles\chris\Application Data\Microsoft\Word\AutoRecovery save of Dickens&Flynn2001.asd Chris BranddC:\WINDOWS\Profiles\chris\Application Data\Microsoft\Word\AutoRecovery save of Dickens&Flynn2001.asd Chris BranddC:\WINDOWS\Profiles\chris\Application Data\Microsoft\Word\AutoRecovery save of Dickens&Flynn2001.asd Chris BranddC:\WINDOWS\Profiles\chris\Application Data\Microsoft\Word\AutoRecovery save of Dickens&Flynn2001.asd Chris Brand(C:\My Documents\07\Dickens&Flynn2001.doc Chris Brand(C:\My Documents\07\Dickens&Flynn2001.doc Chris BranddC:\WINDOWS\Profiles\chris\Application Data\Microsoft\Word\AutoRecovery save of Dickens&Flynn2001.asd Chris BranddC:\WINDOWS\Profiles\chris\Application Data\Microsoft\Word\AutoRecovery save of Dickens&Flynn2001.asd Chris Brand(C:\My Documents\07\Dickens&Flynn2001.docUg@$$d/$$Sg@UnknownG:Times New Roman5Symbol3& :Arial"1hT UF2U+$Z0h2Q~For the past twenty years, the New Zealand political scientist James Flynn has been trying to show that the large (15-IQ-point Chris Brand Chris Brand Oh+'0 ,@LXh |   For the past twenty years, the New Zealand political scientist James Flynn has been trying to show that the large (15-IQ-pointor  Chris BrandhrihriNormalr Chris Brand3riMicrosoft Word 9.0 @^в@"#*:@Z#(2U ՜.+,D՜.+,h hp|  G/+h For the past twenty years, the New Zealand political scientist James Flynn has been trying to show that the large (15-IQ-point Title| 8@ _PID_HLINKSA4y4 8http://www.lrainc.com/swtaboo/taboos/cmurraybga0799.pdf}l'http://www.douance.org/qi/brandtgf.htm}l'http://www.douance.org/qi/brandtgf.htm}l'http://www.douance.org/qi/brandtgf.htm  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]_`abcdeghijklmnoqrstuvwyz{|}~Root Entry F-UA(Data ^1TablefWordDocument&SummaryInformation(pDocumentSummaryInformation8xCompObjjObjectPool-UA(-UA(  FMicrosoft Word Document MSWordDocWord.Document.89q