Social Mobility

Social Mobility Author(s): G. Payne Source: The British Journal of Sociology, Vol. 40, No. 3, Special Issue: Sociology in Britain (Sep., 1989), pp. 471-492 Published by: Wiley on behalf of The London School of Economics and Political Science Stable URL: http://www.jstor.org/stable/591043 . Accessed: 20/09/2014 08:54 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . Wiley and The London School of Economics and Political Science are collaborating with JSTOR to digitize, preserve and extend access to The British Journal of Sociology. http://www.jstor.org This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions G. Payne Social mobility Recentyears have seen the re-emergenceof social mobilityas one of the centralconcernsof Britishsociology,a positionwhich it had lost in the explosionof other sociologicalknowledgein the mid-1960s.In its originalspecialistform, mobilityresearchwas an area in which a small numberof British sociologistsmade significantcontributions. Now, a broader conception of mobility, which capitalises on the effortsof the specialists,has made it a mainstreamtopic for many 1n t ]1S country. more SOC10 Og1StS This is not to argue, as Goldthorpehas done, that in certain respects,mobility 'could well claim a positionof pre-eminence'as a central area of British sociology (Goldthorpeet al. 1980:1;1987:1). Goldthorpe'scriteriafor the claim are the scale of researchprojects, levels of internationalcollaboration,and techniquesof data analysis. The last two correspondto two substantial points of growth, but taken as a whole, the claims of mobility must be more modest. Comparingmobilitywith other topics such as gender,race, work,or sociological theory, we would find many fewer British sociologists involvedin mobilityresearch,fewer (if any?)specialistoptionsin the undergraduatecurriculum,and as we shall see below, an outputgap in termsof new empiricalresearchon the topic beforethe late 1970s. Conversely,these other areas could point to high levels of research investment, international collaboration, and conceptual, if not statistical,elaboration. The high point of researchinvestmentin mobilityresearchwas in the early 1970s, when three majornational surveysof Englandand Wales, Scotland,and Ireland absorbedvery substantialproportions of the SSRC's spending on sociology. Only the Oxford Social Mobility Group retained a sufficientlylarge team of scholars to sustain active work on the data collected, although the number of publicationsthat have subsequentlymade use of the three data-sets must now be severalhundredin number(whichmay go some way to vindicate the SSRC'spolicy against the criticismsof the time). The dismal economicsof large-scalesurveyresearchhave since prevented any similar follow-upsof mobilityper se, but there have been some * * * a TheBritishJournalof SociologyVolume40 Number3 This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions G. Payne 472 smallerspecialiststudies (e.g. Lee 1981;Fiddler 1981) and new data have been derivedfromothersources.These includethe 1983British GeneralElection Study; the OPCS GeneralHouseholdand Labour Force Surveys; the DE Women and Employment Survey; Open University teaching materials (the People in Society exercise);the BritishClass Surveyand a varietyof projectsby the Departmentof Applied Economicsat Cambridge.l As a result, by most conventionalindicatorsof academicsalience, mobilityis thriving.Severalspecialistbookshave appeared(e.g. Dex 1987;Goldthorpeet al. 1987;Halsey et al. 1980;Heath 1981;Hope 1984; Payne 1987a and b); citation scores, in particular of the Nuffield Mobility Study, continue to mount steadily; Council for National AcademicAwards'documentshave shown social mobility as an almost universalcomponentof the first or second year public sector undergraduatecurriculum;recent text books now give more space and more up to date coverage of the topic, and in the last couple of years, the BJS and Sociologyhave carried 14 articles on various aspects of mobility (and about 20, if we include notes and replies, and articles in which mobility makes a brief appearance).2 These dealt wtih gender, elites, technical aspects of measurement, social closure, unemployment,family businesses, language skills, professionalisation,assimilation of migrants, and international comparisonsof class structures. DECONSTRUCTINGMOBILITY Both the range of these articles,and the secondaryanalysisreferred to above, demonstrate how the original paradigm of mobility researchhas becomemodified.Social mobilityis now not so much a single area of sociology as four or arguablyfive connecting clusters of work. The Jirst, not least in terms of paradigmatic dominance, concentrateson discoveringand describinglarge-scale flows of people between social origins and social destinations.It is this which gives rise to our knowledgeabout patternsand rates of mobility; at its heart lies analysis of the mobility table. To accomplishthis, a secondcluster of activity has developed,aimed at refiningthe technical means of ransackingand modelling mobility tables. While no longer an a-theoretical statistical exercise, its discourse tends to emphasise the mechanicsof the process, and is often difficultfor the non-specialistto penetrate.These two clusters can be regardedas mobility researchin the narrowsense. In contrast,the thirdclusteris one which does not regardmobility j7erse a the focus of research, but rather as a subsidiaryprocess which illuminatesthe more importanttopic of socialclass.This, too, has been central to the social mobility paradigm since Glass, particularlyin Britainwhere a numberof theoristshave used data This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility 473 availableas a productof the first cluster to elaboratetheir accounts of the class structure.More recently,however,interesthas grownin using the newer data producedby the national mobility studies of the 1970sand later generalsurveysin afourthway; to explorewider issues, such as occupational and industrial change, the role of women, the fate of migrant groups, or the policy performanceof educationsystems. Finally, drawing on each of the previous four clusters,there is a fifth, explicitly concernedwith the comparative analysisof national systems, in a context of macro-sociologyin its full sense. These clusters are not new. A brief glance at SocialMobilityin Britain(Glass 1954) will show somethingof them at an early stage, and by taking in Lipset and Bendix (1959) and Miller (1960) to cover the comparativeaspect, we would probablyinclude them all. What is new is the extent of the increasedspecialisationof interest and activity (typical of the expansion of all fields of scientific knowledge),the rapid progressmade in each, and as we shall see below, the particulargrowthof the use of mobilitydata to explain a wider range of sociologicalphenomena. Of course, these clusters overlap. The flows which comprise patternsand rates of mobility are measuredby referenceto origins and destinationsdefined by theories of social class. The technical debatestakeon theirsignificancebecausethey empowerthe mobility analyst to address new issues with a more sophisticatedversion of the data from mobility tables. Similarly, individuals have contributed to more than one cluster. Westergaard and Halsey, for example, have separately addressed education and class systems, while Goldthorpehas used mobility to elaborate ideas about the nationalclass structure,helped to promotelog-linearmodellingand more adequate class classificationschemes in current sociological practice, and is presently a leading figure in the comparative analysis of mobility regimes. The purposeof identifyingthese clustersis two-fold.On the one hand, it providesa heuristicstructureto simplifythe descriptionof a complexset of changes.This is particularlynecessaryin the case of social mobility, because its connection with, and dependence on, other areas of sociology raise frequent problems of boundary identification;the interfacewith class theorybeing the most obvious. The clustersalso help to sub-dividethe complexityinto manageable units, allowing for otherwise apparently inconsistent statements about rates of developmentand for explainingthe terms of internal debate. On the other hand, the clusters help to point to a characteristicof social mobilitywhich is unusuallystrong for a subfield of the subject, namely that its core phenomenon can be regardedas either an explanatoror the explanandum.In British sociology,the main motivationbehindmobilityresearchhas been to use it as an explanatorof class. Only in more recentyears has there This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions 474 G. Payne been a growth of a parallel interest in mobility as both explanator and explanandum,in the context of new workon the economy,and on gender. THE GLASSPARADIGMAND THE SECONDGENERATION The orginalcharacterof mobilityresearch,as indicatedby our first two clusters, was set by the LSE Study (Glass 1954), which addressedthe centralissue of the class position of senior managers and professionals,reportingnational rates of mobility for the first time, and developingnew statisticaltechniques.From this point on, mobility research was primarily perceived as involving a large national sample, a formalquestionnaire-basedsurvey, sophisticated computerizedstatistical analysis, and several other specific operational definitions,such as father-to-sonstatus movementsfrom the respondents'originsat the end of compulsoryschooling,the idea of a hierarchyof occupationalclasses, the lay-out of the mobility table, and so on. The centralityof occupationas an indicator,mobilityas percentageratesof flow or specificindices,and educationas a causal factor were also established (not surprisingly in the light of contemporarysociologicaland politicaldevelopments).The interest in comparativeanalysisthat ensuedalso shiftedattentionaway from the occupational distributions that made up the origins and destinations,as these were seen as 'noise' in the comparisonof the actualprocessesof mobilityin differentnations (a key point to which we will return). This rather demanding frameworkprobably helped to inhibit further empirical research:to a large extent, sociologistsfelt they 'knew the answer' to mobility, and additional work would be relatively cost-ineffiective.Indeed, already having evidencethat inequalitiesexisted, that could be used in academic and political debate against those who claimed the death or irrelevanceof class, strengthenedthe professionalsociologistin the face of mereopinion, prejudiceor party ideology. Certainly it is a matter of record in termsof the literaturethat for more than twenty years after 1954, almostall British work on mobility was based on Glass's account, re-usinghis data and core interpretations.3 This was not only the result of a combination of paradigm dominanceflowing from the pivotal positions held by formerLSE sociologists,but also a widespreadconcern with class rather than mobilityper se (our third 'cluster'), and the later anti-empirical tendencyin British sociologyin the 1960s and early 1970s. Indeed, theGlass positionwas not even seriouslychallengedat a theoretical level:as Hope noted in 1974, our ideas This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility 475 far from being derived from some broad body of speculative sociology,have tended to groundthemselvesin an agreedreading of the findingsof the 1949 inquiry. (Hope 1975:1-2) Even the Nuffield Study explicitly set out to follow the younger generationsof Glass'ssurveywho would have becomefathers,rather than sons, by 1972. At the riskof fallinginto the trapof false periodacy,we can see the second half of the 1970sis the key periodin which a new view began to be generally accepted. The publication of Westergaard and Societyin 1975 was both the final Resler's Classin a Capitalist floweringof the previous consensusin its use of Glass's data, and also a major re-statementof the importanceof mobility as a class process.A couple of years later (at the same time as the paperback edition appeared)three key new contributionsto the mobility field became available. One demonstratedbeyond all reasonabledoubt that the data in the Glass mobilitytable could not be trusted,as the table could only arisefroma societyin which therewere virtuallyno class differentialsin fertility, and no historical expansion of nonmanual employment, i.e. conditions absent from all advanced industrialsocieties (Payneetal. 1977).The otherspresentedthe first tranches of data from the new generation of mobility studies, showingradicallydifferentpatternsand rates of movementfrom the 1954 report, and directly challenging the models of class and mobility advanced by Parkin, Westergaard and Resler, and Bottomore(Goldthorpeand Llewellyn1977a, 1977b).This workwas furtherelaboratedin Goldthorpe'ssubsequentbook (1980) and its companion volume (Halsey et al. 1980), while the work from the Cambridgeteam was also beginningto appearin print (e.g. Stewart et al. 1980). In other words, the specialist work of the first two clustersof social mobilityanalysisbecameavailableto a much wider group of other sociologists, not least those in the class/mobility cluster. THE SECONDGENERATIONFINDINGS The first key finding from the 1970s national studies was that of greaterfluidity,over 'long distances',fromlow in the class hierarchy to its upper reaches. The agreed reading of Glass had been that, while there was considerable short-range movement, virtually nobody originatedin the manual workingclasses and ended up in the professional/managerialclass. The Nuffield and Scottish Mobility Studies showed much higher inflow rates of upward mobility, as Table I shows. Detailed comparisons are a little diflicult, because the class categoriesare difFerent,but whereas 52 This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions 476 TABLE I: G. Payne % InJ%ow mobilityintotheuppermiddleclassa OriginClass Glass A B C D E F 47.7 11.7 12.6 15.3 2.3 e 18.7 1.1S (Managers/Professionals) (IntermediateWhiteCollar) (RoutineWhite Collar) (SkilledManual) (Semi SkilledManual) (UnskilledManual) Nuffield 24.2 12.5 35.0 15.7 ll2 6 s 28.3 J * J SMS 23.3 19.6 21.3 16.5 9.3 s 35.8 1OOJ 100% 100% 100% (n = 262) (n = 1285) (n = 550) Notes: * The classesA - F have been createdby takingthe equivalentoriginalclassesfromeach of the studies:this worksquite well for the uppermiddleand manualclasses,but less well for the lower middle sector. The followingshows the details of the groupingof the studies' originalclass categories: Glass: A = I+II; B = III; C = IV+Va; D = Vb; E = VI; F = VII Nuffield: A = I; B = II; C = III,IV+V; D= VI; E+F= VII SMS: A- I; B = I; C = III+IV; D = V; E = VI; F = VII a Adaptedfrom:Miller 1960:71;Goldthorpe1987:45;Payne 1987b:65 per cent of Glass's upper middle class was upwardlymobile, with only 19 per cent fromthe manualclass, the NuffieldStudyshows 76 per cent upwardlymobile, with more than 10 per cent comingfrom each of the other classes,including28 per cent from the two manual classes, a level slightlyexceededin the Scottishdata. It was this new evidence that challenged existing models of class boundariesand closure. We can also comparemobilityfrommanualto non-manualclasses (using the Kelsall/Miller version of Glass's data) in the three studies, as well as, more cautiously because of the different categories,gross mobility. This is shown in Table II. We again apparentlysee more fluidityin the more recentstudies, and the second new finding that while upward mobility can apparentlyincreaseor be high, downwardmobilitydoes not need to increaseor to be equallyhigh. The new occupationalopportunityin the non-manualclasses createsconditionsin which the sons of nonmanual workersneed not be displaced to accommodateincomers from below. Mobility is not a zero-sum game. The low rates of downwardmobility may explain lack of interest in this aspect of mobility (Richardson1977, excepted). This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility TABLE II: 477 Mobilityratesin 3 studies Glass Nufld SMS 15.6 15.6 21.5 47.3 23.0 13.8 32.3 30.9 23^0 11.3 23.2 42.5 28.2 33.1 38.7 40.0 25.6 34.4 38.4 26.5 35.1 Manual/Non-Manual Mobility: - Upward Downward ImmobileNon-Manual ImmobileManual GrogMobility - Upward - Downward - Immobility Notes * Calculatedover classes A7 B, C, D, and tE + F': see Table II above for derivation.The numberof classesaltersthe mobilitymeasured(the greaterthe numberof classes,the higher the apparentmobility).The readeris remindedthat these rates are broadindicatorsonlyS becausethe classesare only equivalentsacrossthe 3 studies. The second generationstudies would have provided a historical pictureof trends had the Glass data been reliable;instead they had to rely on 'internal evidence from a comparison of cohorts at differentstagesof theircareerdevelopmentto identifychangingrates of mobility. Goldthorpehas shown how absolute rates of mobility indicate that younger men have better chances of obtaining nonmanualand serviceclass positionsthan older men (Goldthorpeet al. 1987). On the other hand relativerates do not improve (see below) while the Scottish Study has shown that for access to non-manual jobsnparticularlyon first entry to the labourmarket the majorgains in improvingthe opportunitiesfrom sons of manual workerswere made by the late 1950s and have hardly improved since (a point confirmedby the BritishGeneral Election Surveydata: Goldthorpe et 1. 1987:262,Payne 1987b). It could thereforebe claimedthat in termsof the key questionsof 'how much mobility is there?7and Cisit increasingor decreasing?, the main studies have now provideda substantialanswer.4The data in a processedand publishedformare availablefor other sociologists to use, and with the up-datework to 1983fromthe GeneralElection Survey ((;oldthorpeet al. 1987) showing no radical departurefrom the 1970s picture)a fairly comprehensivegap in our knowledgehas been filled. Of course, this basic increase in informationhas gone hand in hand with conceptualdevelopment,and it is here that these This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions 478 G. Payne apparently straightforwardanswers and steps forward begin to . c .lsappear . agaln. DIFFERENCESIN APPROACHES Perhapseven the suggestionabove that there were straightforward answers was over-optimistic.The reports from the mainstream studies require careful reading, as their arguments and use of evidenceare dense. Even apparentlysimple things such as counts as mobilityare hedged in by detailed operationalisation. Goldthorpe'smain analysisof mobilitycoincidedwith his interest in the 'serviceclass' and doubts about a neat hierarchicalorder to the class structure.There is naturallyan emphasis on the service class in his subsequentwriting,and his generaldiscussion(although not his tables of data) tend to 'speakof upwardmobilityonly in the case of movementinto classes I and II, ratherbetween the other 5 classes' (Goldthorpeet al. 1980:42).An even moreimportantfeature is his focus on relative mobility between people from different origins, rather than using an absolute count of how many people were mobile.This enableshim to differentiatebetweenmobilitydue to changes in occupational distributions, and changes in the processesof class inequalities.It also leads to the elegant paradox that while, absolutely,more people are upwardlymobile, the extra supply of non-manualjobs equally advantagesthose already born into the non-manual classes, so that the relative chances do not improvefor the sons of manualworkers.For some people, this is the key findingfrom the 1970s. The modelof'constant social fluidity'in this relativesense can be used to point to a failureof post-warwelfarereformsor class politics to reduce class differentials in inherited life chances. Given Goldthorpe'scentral concernwith how mobilityrelates to the class structure,it followsthat a pessimisticview of relativemobilitylooms large in the sophisticateddeploymentof odds-ratioand log-linear modellingaroundwhich much of his workhas been built. While he recognisesthe role of occupationalchangein total mobility,he is not particularlyinterestedin it precisely because the interest in total mobilityis 'the resultof changesin objectivemobilityopportunities, and did not reflect any changes ... in the direction of greater equalityof opportunityor "openness"'(Goldthorpeet al. 1987:253).5 The patternof class inequalityhas been confrmed, ratherthan modified by occupationaland mobilitychanges (see Payne 1987c). In contrast,the approachof the ScottishMobilityStudy has been to emphasise the occupational character of social mobility, by charting industrialand occupationaltransition,and attributingto employment(which definesthe actual origins and destinationsused This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility 479 in the mainstreamstudies) a causal power in determiningmobility rates. This view is not so much a negationof the importanceof class an an attempt to treat it in a series of more specificforms:mobility as a productof class is shiftedto class as being partiallya productof mobility, the latter in turn being an outcome of labour market processes and the deployment of capital. Combined with a more conventionalview of occupationalclasses as a hierarchy,this leads to concentrationon absolute mobility and in particularinflows to current classes, rather than relative outflows from origins (Clarke, Modgil and Modgil 1990). While not optimistic about changing class inequalities, the conclusion is that these are not so much confirmed,as modified,by structuralchange. On the surface,the contrastbetween the two approachescan be read as a differencebetween specialistsin the mobility field, or as alternativemethodsof statisticalanalysis:that is, to say, issues that are really mainly the concernof those workingin the first two of our clusters. However, at the heart of the two perspectives lies the problem of how mobility relates to class analysis. Goldthorpe, despite his new model of the class structureand his methological contribution, speaks for the tradition that constitutes the third cluster of mobility research,that of mobility as a class process. As his choice of book and chaptertitles signal, he is interestedin 'class mobility' not 'social mobility', the same interest which has driven most of the many British'consumers'of data on mobilityin the past. This perspective,in most of the work done up to 1980, treated mobility as being tied to structural class analysis. Given the assumptionthat occupationalstatus could be convenientlyused to denote individual class position, this discouragedreflectionabout operationalisations,and gave a particular character to our perception of mobility. A second strand of writing, in the context of a British sociologyin which class has always been central, has made use of mobilityin a looser way as part of more descriptiveaccounts of'who gets what' in Britain today. One could argue for example, that research on cycles of deprivation, second generation black immigrants,or much of the debate about equalityof opportunityin education, comes under this rubric: the line between 'mobility research'and researchin which a broad mobilityapproachis used, is a blurredone. This kindof usage relegatesmobilityto a subsidiary position:in both cases, mobilityhas been as strongor as weak as the core class analysisat the time. Mobilitywas kept on the agenda by our interestin class, but remainedessentiallyan item requiringlittle discussion in its own right. A recent example is the discussion of proletarisationand associated issues in the British Class Survey (Marshallet al. 1988). Thus writerson class structure,in particularsuch as Bottomore, Miliband, Westergaard,Parkin, and Giddens6have re-interpreted This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions G. Payne 480 mobilityfindingsto amplifytheirown arguments.This has certainly kept mobilitya live sociologicalissue, but only in the narrow context of class: in this view the study of social mobilityis usefulin illuminatingthe of the social stratificationsystem in capitalist societies,operation but it is essentiallysubordinateto the real stuffofclass analysis.(Kelsallet al. 1984:116) Indeed, from a Marxist perspective,mobility (and its study) are undesirable, focusing on the individual rather than on social structure,and creatingfalse consciousness:the 'blinkeredchase for the elusive carrot'preventsthe donkeyfrom recognising that others are similarlyplaced (ibid.). The 'blinkeredchase' analogy can equally well be applied to the consequencesfor social mobility of the pursuit of class analysis. Becauseit had the status of explanator, mobilityper se was left for long as a relativelyundifferentiatedconcept (except, perhaps too in the fieldof educationand mobility),to be used only in one particular set of arguments.We can illustrate some of the consequencesof this narrowfocus with two examples. First, it is a commonplacecriticismof mobilitystudies that they ignorewomen.7 One reason for this tendency is the difficulty makingsense of the class positionof women, which is what we of do because the mainsteam focus has been on mobility must as class analysis.Class positionis for men given by their normal occupation: somewomen have never worked,marriedwomen commonly take at leasta break from employmentduring the family building stage, some then work part-time, and others never return to paid employment.What then is the class position of these women, and hencehow have they been mobile? The recent and continuing debatesabout women and social class have contested the virtuesof treatingthe wife as having her own class, her relative class,a mergedpartnershipclass, and a dominantclass husband's onthe groundsof full-timeratherthan part-timeand calculated higher rather than lower,employment,(e.g. see Abbottand Sapsfordstatus Thechoice involvesa decisioncentralto ideas of mobility, 1987). theunit of stratificationthe individualor the household?namelyis The failure to develop a mobility of women derives then from failure of class conceptualisation.Had mobilityand class not been a so intimatelyrelated, then a female mobility might have emerged earlier, and indeedwould have reinforcedthe need for an alternative perspective on female social class. Had we asked in a less focused waywhat is it in a woman's life experiencethat she (and others) perceive as major changes in social identity, then we would have confronted our narrow occupational class definition of mobility muchsooner. For example does marriage provide a medium for This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility 481 mobility?Is the father-daughterdefinitionof mobility'sorigin and destination the best way to account for movements, when men participate in labour market and class processes in a genderdifferentiatedway? Would not mother-daughtertransitionsbe more interesting,particularlyfor exploringtrendsin life experiencerelated to new patternsof educationand labour marketparticipation?Most importantof all, we would probablyhave grapsedhow narrowwas our operationalisationof mobility, simply because the mobility experiencefor women is different.As Sorokinarguedmore than half a centuryago, there are many channelsfor upwardmobility,even if the main dimensionis read as that of class:wealth, influence,status and marriageare independentspheresof mobility (1927:133-80). So too is employment.Gender segregationin the labour market constrainsinitial opportunityfor women (Hakim 1979;Payne et al. 1980). Downward mobility occurs later in life, as Dex ( 1987) cogently shows: re-entry to paid employmentfor marriedwomen with young childrenis typicallyon a part-timebasis, and at a lower level than beforechild-bearing.The totalityof the woman'ssituation explains her occupationalcareer:mobilityis not uni-dimensional. Perhapsthe best example is the work of the Essex group on the British Class Survey (Marshall et al. 1988). Although in other respects firmly centred on class analysis, their treatmentof female mobilityleads them to distance themselvesfrom Goldthorpe'smore traditionalpositionon the unit of mobilityanalysis.While accepting his view that socio-political class formation needs to take little account of female careers, their interest in demographic class formationand the distinctiveworkexperienceof womendirectsthem towards absolute measures of mobility, structuralchange, gender segregatedlabour marketsand the wider contextsof women's lives. This in turn feeds back into their overall conclusions about contemporaryclass. The second case of excessivelynarrowfocus is that of the Marxist view of mobilityas bourgeoismystification.Goldthorpehas seen this view as a central block to a proper understandingof the mobility process, and attempts to legitimatehis own position at the start of SocialMobilityand ClassStructure in ModernBritainby, among other things, demonstratingthat Marx dealt with mobility. However, in doing so, Goldthorpeaddresses the problem of individual movements across a class structure,thus reinforcingthe way that it has been read by most Marxists. A closer examinationof the actual quotations which Goldthorpe uses shows that Marx wrote more about whole categories of persons experiencingstructural mobility. Much of his accountof the emergenceof capitalismis dependenton changes in the types of employersand employees. The growth of drittepersonnen, of bureaucratsand intellectuals,and the declineof the petty bourgeoisie are changes to both class and occupational This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions 482 G. Payne structures, requiring changes in circumstancesfor large sets of personnel. This points away from the narrow focus on class membershiptowardsa broaderhistoricaland economicframework. In other words, there is space even in the Marxist canon for an elaboratedview of mobility, even if it is not one that is concerned with individual movements.8 The Marxists' general antipathy towards mobility research can be seen to be in part due to this restricted conception of mobility which obscures the broader conceptionof structuralmobility as a featureof capitalistsociety. MOBILITYAND OPPORTUNITY In both of these examples, mobility has had the logical status of explanator of class mobility.For a long time, the only substantialarea of work in which mobilitybecame,at least partially,an explanandum was the link betweenschoolingand career,althougheven here, two of the main theoretical contributions both treat education as basically only one link in a complex chain of class reproduction. Giddens sees educationalqualificationsas the key determinantin marketcapacity,and thereforemobilitychancesassumea significant position in the class structuration process (Giddens 1973:107; 181-6).9Parkin,on the other hand, in shiftingthe debate away from 'inequalities surrounding the occupational order', places even greateremphasison credentialismas a strategyof social closurefor the middleclass. While his positionallows for an analysisof closure and movementbetweenother differentiatedgroups (on the grounds of language, race, gender, or religion) this is not developed, and mobility remains in subordinateposition in his argument (Parkin 1979; Payne 1987a:72-5). Both Giddens and Parkin do, none the less, begin to open up possibilitiesfor an elaborationof mobility as an interestingprocess in its own right. To some extent this had alreadybegun to emerge elsewhere in the study of education and mobility. Following the initial interest in the consequences of the 1944 Education Act reported in Glass (1954), numerous attempts have been made to explore preciselywhat parts of the educationalprocess promoteor hinder successful employment outcomes. The underlying concern may still be, as Halsey has said, a class one: to what extent is Britain an 'open society'? (Halsey 1979). Nevertheless, the research following from that question has investigated the likely causes of mobilityexperiencein a way that has helped to unpackthe idea of mobility itself. It is at this point that our third cluster of mobility researchbegins to mergeinto the fourth,with its wider perspective. The educationalprocess can be sub-dividedinto three elements: what factorsin family of origin promotesuccess in education;what This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility 483 featureof school systems or events in schoolingshape qualification profiles;what do employersactuallydo when they select indivi'duals fromthe generalsupplyof qualifiedmanpower?In the first,parental cultural capital has been identifiedin a number of ways, from the stimulus of intellectual artifacts in the home, through direct attitudinal encouragement,to selection of particular schools or financial'investment.Not only are these relevant factors, but they vary over time: work at the Centre for Educational Studies in Edinburghhas recently linked parents' levels of education to the propensity of their children to qualify for higher education, and drawn attention to the rapid, historically-specificrise in education levels among the currentgenerationsof parents (THES 1989a). The way in which family backgroundinteracts with the school system has been extensivelyexplored.The emphasisin Halsey et al. (1980), which has examined IQ, primary schooling, secondary schooling and higher education in some detail, has been on origin and educationaldestination,but the overall link between education and occupationshas also been shown by Halsey (e.g. 1979). Early optimism ('given the diminishingimportanceof economicand social backgroundsas a determinantof the type of secondaryeducationa child receives,social mobilitywill increase':Glass 1954:24)has thus been succeeded first by demonstrationsof the resilience of class differentials in the face of educational reform (e.g. Little and Westergaard1964, but also see Gray et al. 1982 and the case of comprehensivesecondaryeducationin Scotland) and then by more systematicdisaggregationof the educationalprocess. Other studies have extendededucation,from schooling,to include post-secondary education (e.g. Psacharopoulos1977; Raffe 1979; Blackburnet al. 1980)in orderto fill the gap left by the morelimiteddefinitionsused in earlierwork.'° At the risk of colonisingother parts of the educationfield," it can also be said that we know a lot more about the social processes within schools which determinequalificationoutcome. The impact of teacher expectations, the influence of wider culture and peer group, and socialisationinto acceptable aspirationshave all been documented.To give but one illustration,early subject'choice'cuts off specificoccupationalavenues,or even whole classes of career:as the Youth Cohort Study, Sweep Two shows, gender differences become reflected in choices of post-secondarycourses and institutions well beforethe end of compulsoryschooling (THES 1989b). The third stage of the progressionfrom infant to employeeis the crucialstage of employerselection.Again, early optimismabout the meritocraticpossibilitiesof job access through qualificationshave been replacedby recognitionthat the childrenof the middle classes do well in the educationalrace, but that formalqualificationsare not all they seem. This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions G. Payne 484 It could be arguedthat the associationbetweenqualificationsand job success comes about not because employers care for qualificationsthemselves but because employers choose young peoplewith favourablebackgrounds,attitudesand behaviour,and that these young people also tend to have good qualifications. (Gray et al. 1982) In this view, qualifications are an indicator not of intellectual achievement,or learnedskills ready for employment,but ratherof more general attributes, of personality, motivation, manners, or culture.Some of these things may be of direct relevanceto work time-keeping,application,problem-solving,acceptance but others reflectthe preferenceof employersfor 'peoplelike us'. At the higher levels, this constitutesclass preference,which may operateapparently impersonally,or directly(see Kelsall 1974). Reid, in his discussion of the school/worktransition,gives considerableweight to this, not least in showing low levels of employer knowledge about the educationthat lies behind the qualification(Reid 1986). MOBILITYAND EMPLOYMENT Employersrequiresome mechanismsto enable them to select from among many applicantsfor employment.In the earlierpart of this century, relativelyfew people had anything other than elementary education, while more firms were small and family-controlled. Selectionfor the betterjobs could be and indeed had to be-on the basis of personalknowledgeor by referenceto the type of school and family background,ratherthan the rarerformalqualifications. Even more to the mobility point, the senior positions tended to be filled by the owners' offspring who would directly inherit the company as propertywhen their fathers died.l2 The decline of the family firm coincidedwith an expansionin the supply of education, from which the 'crown princes' also benefitted,but on which they did not depend. Ratherthan describingthe expansionof education as a shift towardsuniversalisticvalues it would seem more accurate to view it as a new mechanism performing the old function of social reproduction. Social inheritance, whether through the transmissionof property or through the transmission of cultural capital, is still social inheritance.(Karabaland Halsey 1977:19) This view has gained ground over earlier expectations of meritocraticopenness, althoughas Parkinhas argued, middle-class strategies of credentialistclosure may work against some of the children of the professional middle classes who are less bright This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility 485 (Parkin1979:61). But on the whole, the formal education qualificationsrequired'as professionalisation,bureaucratisationand automationof work proceed' (Little and Westergaard1964:302)have beenmore easily obtained by the middle classes. It is important to note how this educational process occurs alongside,but not at the same rate as, occupationalchange. The for 'technically' qualified manpower has increased at the demand sametime that the supplyhas been increasing.There is, however,no natural'match' between the supply of highly qualified people and thenumberof posts that need ideally to be filled with highly qualifiedmanpower.With an overall pattern of demand exceeding supply, some posts have to be filled by people lacking formal qualifications.There remain opportunities for non-credentialist mobility,and the associationbetweeneducationand employmentis therebyattenuated. Furthermore,we need to consider how that demand has been constituted. In particular, the shift of employment into large, bureaucratic,and often public sector organisations,and away from small manufacturingcompanies has created a distinctive demand. The processof occupationaltransitionis not a uniformincrease in theproportionof non-manualjobs. Rather,it consistsof the creation of certaintype of new employmentin specificindustrialsectors.The broadincreaseacrossall sectorsbetweenthe Warshas been replaced by a growth of employmentrelativelymore concentratedin larger organisations,and in the 'newer'parts of the servicesector, such as the welfarestate, financeand governmentagencies.These industries have historicallyused credentialismas a mechanismfor recruitment, so that their expansionhas createdjob opportunitiesfor those with credentials,not least the sons of the middle class. The declining sectors, such as manufacturing,have a smaller work force, and so their traditionallyhigher rates of upward mobility are now a less significantfactor in the total picture (Payne 1987b:122-54). This view of mobilitycharacterisesmuch of the work in what we have called the fourthclusterof mobilityanalysis.Here, althoughan of (class) mobilityare interestin class is not far away, the explanations being sought in educational and occupational processes. This emphasis on occupational mobility, and hence on labour market process, industrialisationand de-industrialisationis strongly representedin the workof the ScottishMobilityStudy. While at one level this approachcan be regardedsimply as a disaggregationof class processes,the crucialdistinctionis an interestin mobilityin its own right, and the wish to account for its specific forms in an occupationalframework.This is clearly shown in such examplesas the explorationof short term job changes (Dale et al. 1984); the investigationof local labourmarkets,the careersof clerks,or types of work (Blackburn and Mann 1979; Prandy et al. 1982); or This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions G. Payne 486 de-industrialisation(Lovattand Ham 1984). It is also prominentin recent issues of the BJS: see for instance the articles by Savage, Green, Bonney, and the debate between Cassis and Chapman, in Volume 39. The same volume also includes an example of an associated area of work which links mobility into the analysis of other, less narrowlyclass-based,topics, namely Hornsby-Smithand Dale's explorationof the assimilationof Irish immigrantsin England (1988). The broader interest is able to draw on a mobility perspectiveto provideone elementin the total picture.Other recent cases are Hutson's account of the farming industry (1987) and Evans's work on language (1987), or the new body of work on women's paid employment(Dex 1987;Chapman 1989). COMPARATIVEPATTERNSOF MOBILITY In contrast,much of the comparativeanalysisof mobility (our fifth and final cluster) has proceededin a more traditionalway. This probably reflects the dominance of American sociology and in particularthe development,from Lipset and Zetterberg'swork, of the Featherman,Jones and Hauser thesis of underlying crossnational similarity in relative mobility in countrieswith a market economy and nuclear families. A small number of British sociologists, such as Hope (1982) and Breen (1987) have contributedto the associated debate about modelling the mobility process for crossnational comparisons, but the main strengths of this country's sociology have not lain in statistical analysis. Indeed, without Goldthorpe's efforts in collaborationand through the CASMIN (ComparativeAnalysis of Social Mobility in Industrial Nations) projectat Mannheim,there would be little to show in this, our fifth cluster. Given what was said earlier about Goldthorpe'spreferencefor relativemobilitymeasures,it will not be surprisingto relatethat the bulk of the CASMIN output has focusedon this approach.Using a recodingof an increasingnumberof national data sets, the project has identifieda commonality,though not a uniformity,of relative mobility rates. Within the range of countries, Britain (or more strictly, England and Wales) falls near the centre of possible outcomes. Equally,given the preferencesof the presentauthor,it will be no surpriseif more attention is given here to what CASMIN tells us about absolutemobility rates. The occupational distinctivenessof Britain is its low proportionof employmentin agriculture,and the high proportionin the industrialmanual sector. It follows that the Britishserviceclass is recruitedmorefromthe latter, than is the case in other Europeancountries.Conversely,as a result of our earlier This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions 487 Socialmobility industrialisation,we have fewerupwardlymobileor manualworkers recruitedfromfarmingstock. Whereason these indicators,Britainis top (or bottom)of the 'league',in self-recruitmentto the serviceclass and manual class neither England and Wales nor Scotland is distinctive(Goldthorpeet al. 1987). Both absoluteand relativemobilitymeasuressuggest that there is a core social fluidity, as might be expected from the similarity of industrial structuraldevelopmentidentified by convergencetheorists, but not a uniformityof occupationalor class distributionsand processes (see Erikson et al. 1979, 1982, 1983; Erikson and Goldthorpe1987). A longer historicalview, using a wider range of smaller studies, has none the less shown more variety in mobility patterns,and a weakerfit betweenthe industrialisationand mobility factors,combinedwith increasingsimilarityin the business elite of Britain,America,France,Germanyand Sweden (Kaelble 1986). As these four other countries have been identified as displaying characteristicdifferencesin other partsof their mobilityregimes,the next step seems to require more detailed and concrete analyses of institutionsand processes,much as Muller has recently attempted for educational'cultures'(1987). MOVINGFORWARD The comparativeanalysis of mobility suggests several possibilities for future researchin this country. If we are unusual in the size of our manual workingclass, this not only createsdifferentpatternsof mobility flows which are worth studying but also indicates that the process of de-industrialisationrequiresparticularattention. To be morespecific,what are the mobilityexperiencesof employeeswhose industriescontract,leaving them geographicallyand occupationally isolated in a shrinkinglabour market.Workon formersteelworkers has begun to show some of the effects in terms of family life and would-be small businessmen,and the ESRC EconomicLife project has also producedsome data (see Harris et al. 1985). Perhapsit is even now too early to assess the full impact. Second, if there are distinctive'credentialistcultures',an examination of how these changeover time would seem to be long overdue. The growth of private secondaryeducationin the last decade, the 'flightfrom the inner city' with its attendantproblemsof schooling, the new sixth form and FurtherEducationColleges, the rise in HE applicants from Registrar General's Social Class II, the new qualificationsavailable in HE, the present Government'spolicy on HE and its obsession with training engineers,and the major new role of the Polytechnicsin educatinghalf of our graduates,all cry out for both systematicand contemporaryanalysis. The supply side of This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions G. Payne 488 qualifiedmanpowerhas not only changedsince the nationalstudies of the early 1970s, but continues to change. How it relates to the changing demand side is actually becoming more opaque as both factorsmove indepently. A special case within this area is the position of young people. Youth unemployment,youth trainingprogrammesof variouskinds, and a changinglabour markethave set up new conditionsfor early life experienceof mobility.Again, it may be too early to see if this representsa 'blip' on the normalpatternof class reproduction,or a significantnew development:certainly it is only recently that we have been able to see some results from the sociologicalstudies of youth that investigatedchanges at this end of the market. Mobility studies have had little to say about early work life experience:following census coding conventions,the tradition has been to ignore'temporary'jobs beforethe first 'realjob', and to treat apprenticeshipsas functionallyequivalent to skilled manual work. However,the young man who did a year as a butcher'sdeliveryboy beforegoing into the shipyardto servehis time as a turnerand setter (to take a characteristicmale examplefroma previousgeneration)in fact went throughtwo stages of status change by the age of 21; from casual unskilled boy to apprentice,and from dogsbody trainee to skilled man providedof course he was not laid off as soon as he qualifiedfor a skilled man's wage, when he experienceddownward mobility. If mobility is about class behaviour, a whole area of experiencehas been virtuallyignored. The same is true at the other end of the life-cycle.Retirementis not only a traumaticstatus passage,but one which changespatterns of association, calls for re-assessmentof self-identity,and causes financialdislocation.The traditionof treatingold people as having the class of theirformeroccupationmakesa nonsenseof understanding the mobilityexperiencesthat go on afterthe age of 64, the age at which the majorstudiesconvenientlycut off. Of course,if mobilityis treatednarrowlyas class mobility,as signifiedby position of a fulltime paid job, then the elderly are defined out. This seems doubly unsatisfactorywhen the retired are already 30 per cent of the population,and will increasinglyhave a longer life. This slightlyembarrassingintrusionof demographyalso bringsus again to the problemof supply and demand.Some of the problemsof youth employment in the 1980s was due to the steady rise in numbersof school leavers coming onto the labour market.Equally, someof the fall in the numbersof unemployedpeople is due to the sharpdeclinein these numbers.Majoremployers,such as the Army, the banks,and the Health Serviceare eithertalkingabout changing, or have already changed, their recruitment policies, to access differentgroupsor to securetheir earlierpositionagainstnew rivals. These changes come on top of the extensive changes to industfial This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility 489 and commercialoperationsbased on new technology,which again have only been partiallyresearchedfrom a mobility perspective. While total mobilitygoes beyond initialjob entry, later outcomes are associatedwith entry points, and both are-or rather, in our sociology, should be related to the real life events that shape employmentexperiences.On the one hand, there is slump, war, long-termunemployment,and the birth of a new underclassas we have belatedly come to recognise (Payne 1987a; Goldthorpeand Payne 1986a). On the other is the new economy of the south-east, the generationwhich is inheritingdomestic propertyas the norm, the shift to self-employment,and the 'new' establishment(Observer 1989). Where is the serious sociology of the Yuppie (do we remember what it stands for?)? It is missing precisely because mobility studies have failed to recognise that mobility is not just about movementacrossjobs, but also, interalia, movementsbetween levels of wealth and income. When a formerConservativeCabinet Ministercan call his autobiographyUpwardly Mobile,surely the time has come to recapturemobility, not just from the politicians, but from the intellectualghetto of a narrowclass orientation.We shall only achieve the goals set for us by those who have sought to understandclass processes,by first expandingour understandingof the full complexityof the conceptof mobility,and then by analysing our own society and lives to see precisely how this complexity is manifestedand remanifestedin contemporaryBritain. G. Payne Facultyof SocialScience SouthWestPolytechnic NOTES 1. For examples, see respectively Goldthorpe and Payne (1986a); the work of the Stratification and Employment Group at Surrey University, such as Gilbert (1986); Martin and Roberts (1984) and Dex (1987); Abbott and Sapsford (1987); Marshall et al. ( 1988) and Steward, Prandy and Blackburn(1980). This is by no means an exclusivelist, but it is not the intention of this article to presentan annotatedbibliography.The authorknows of no currentlisting, but has found Mack et al. ( 1957), Bibby (1975) and Kaelbe (1981) usefulon the older material. If, because of the particularargumentsin this article,any fellow mobilityanalystsare not named, I hope they will acceptmy apologiesfor the omissionof their contributions. 2. This briefcount coveredarticlesin the two major Britishjournals during 1987and 1988.Cautionshouldbe taken about such small samples:the count for a single 'journal-year'rangedfrom one to nine in the two successive years. Several of the contributorswere not permanentlybased in Britain, and not all the articles dealt exclusively with Britishsociety. 3. Despite the contributions of Benjamin(1958) and Noble (1972), or Runciman ( 1966) and Richardson (1977), other writersfrom text booksto monographsdrawon Glass:see Worsley et al. (1977), Giddens (1973), and for a This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions G. Payne 490 longer account,Payne (1987a). 4. This article can only select a few resultsfromthe severalbooksand many papers which have resulted from the mainstream studies. In making that selection, the intention has been to simplify the issues and make them accessibleto the sociologistwhose main interestdoes not lie in this field. Heath's SocialMobility,althoughnow a little out of date, is a usefulprimerto the Nuffield Study. 5. I t follows that the account of mobility as presented in the previous section would very probablynot be the way in which John Goldthorpewould preferto presentit. 6. Otherswho have writtenon class, or reported the debate in general discussion, include Bilton, Frankel, Kelsall, Marshall, Musgrove, Raynor, Swift and Worsley. 7. Although the criticism is largely true,it is in partmisplaced.Glassdid in fact collectfemaledata, but this was not analysed. We have already noted the gap in researchin the 1950s and 60s which applies to men and women. One contribution in this period was the Labour Mobility study by Harris and Clausen ( 1967) which did include women. While the Nuffield Study did not, the IrishStudydid (see forexample Hayes ( 1987) and the Scottish Study interviewed wives of the male respondents. 8. A case in pointwhich reliesless on the structural and more on the processualis the Polish work of Mach and Wesolowksi,recentlytranslatedinto English, and which draws heavily on aspects of work by Britishwriterssuch as Parkinand Giddens. 9. Giddens's model of three main classes is largely dependent on the assumption that educational qualificationsare broadlystratifiedinto three levels, and that mobilitypatterns also reflectthis division.The latterowes a greatdeal to his use of the Glassdata, with its apparentabsenceof long-range mobility(see Giddens,181-6). A similar dependenceon the 1949 study can be found Parkin(1979). 10. School-centredsociologiststended to ignore 'on the job training'. Apprentieeship,for example, was an educationalexperiencefor well over one thirdof men (at one stage, 45% of male Scottish school leavers took apprenticeship),but which has largely been ignored both as a mobility generator,and as a mobilityexperience. 11. As noted earlier, mobility has several interfaees with other subjeet areas, where the exact boundariesare somewhat blurred. It is preeisely at these boundariesthat some of the most fruitful work ean develop, and it is unimportant whether we label the currentdiscussion'mobility',or gender, or raee, or edueation. 12. Even in the more recent studies, ownersof small businessesand farmers were found still to have distinctively high ratesof reeruitmentfromfathersof the same occupation. In the Scottish Mobility Study, two thirds of farmers and one thirdof businessmen wereselfrecruitedin this way, comparedwith a rate of 5% self recruitmentin the rest of occupations with similar HopeGoldthorpescale scores. This is all the more strikingas the surveyshowedless than half as manyjob opportunitiesfor sons in farmingand businessesas found among the fathers'generation. BIBLIOGRAPHY Abbott, P. and Sapsford, R. 1987. Bibby, J. 1975. 'Methodsof Measuring Women and Social Class. London: Mobility'.Qualityand Quantity9(2). Blackburn, R. and Mann, M. 1979. Tavistock. Benjamin, B. 1958.'Inter-Generational The WorkingClass in the LabourMarket. Differences in Occupation'. Population London:Macmillan. Blackburn, R. Stewart, A. and StudiesXI. This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions Socialmobility 491 Prandy, K. 1980. 'Part-timeeducation Glass, D. V. (ed.) 1954.SocialMobilitvin 14. Britain.London:Routledgeand Kegan route'.Sociologv and the "alternative" Paul. Household 'Gender, 1988. N. Bonney, Goldthorpe, J. et al. 1980, 1987. Social and Social Class'. BJS 39(1). Breen, R. 1985. 'A Frameworkfor the Mobilitvand Class Strueturein Modern Comparative Analysis of Social Britain.(lst and 2nd editions), Oxford, ClarendonPress. 19(1). Mobility'.Sociology Breen, R. 1987. 'Sources of Cross- Goldthorpe, J. and Llewellyn, C. National Variation in Mobility 1977a. 'Class Mobility in Modern II(2). Britain'.Sociologv 21(1). Regimes'.Sociologv Cassis, Y. 'MerchantBankersand City Goldthorpe, J. and Llewellyn, C. 1977b. 'Class Mobility: Illter-generaAristocracy'.BJS 39(1). Chapman, A. 'Gender and graduate tional and W ork-I,ife Patterns'. BJS Paperread at BSA 28(3). under-employment'. Annual Conference,Plymouth, March Goldthorpe, J. and Payne, C. 1986a. 'Trends in Intergenerational Class 1989. Mobility in England and Wales Youssef to 'Reply D. 1988 S. Chapman, 20(1). 1972-1983'.Sociologv Cassels'.BJS 39(1). Clark, J. Modgil, S. and Modgil, C. Gray,J. McPherson, A. and Raffe, D. Education. of Secondary . John 1982.Reconstructions (eds) 1990 (forthcoming) Goldthorpe:Consensusand Controversy.London:Routledgeand Kegan Paul. Green, S. D.J. 1988. 'Is Equality of Brighton:Falmer. 's OccupationalOpportunitya False Ideal for Society'. Dex, S. 1987. Women BJS 39(3). Mobility.London:Macmillan. Segregation. Dale, A. Gilbert, N. and Arber, S. Hakim, C. 1979.Occupational 1984. OccupationalClasses and Class Departmentof EmploymentResearch. Switches.Stratificationand Employment PaperNo.9. London:DoE. Group WorkingPaper, Departmentof Halsey, A. H. 1979. 'Social Mobility Sociology, University of Surrey, and Eucation' in D. Rubinstein (ed.) Equality. Education and 1979. Guildford. Penguin. Erikson, R. and Goldthorpe, J. 1987. Harmondsworth: 'Commonalityand Variation in Social Halsey, A. H. et al. 1980. Originsand Oxford:ClarendonPress. Fluidity in IndustrialNations Parts I Destinations. Harris, A. and Clausen, R. 1967. Review3. Sociological and II'. European Erikson, R. Goldthorpe, J. and LabourMobilitvin GreatBritain.London: 1979. 'Inter- HMSO. Portocarero, L. generationalMobilityin Three Western Harris, C. C. et al. 1985. 'Redundancy in Steel' in B. Robertset al. 1985. New EuropeanSocieties'.BJS 30. Life. Manchester: to Economic Erikson, R. Goldthorpe, J. and Approaches Portocarero, L. 1982. 'Social Fluidity ManchesterUniversityPress. Hayes, B. 1987. 'Female Interin IndustrialNations'.BJS 33. Erikson, R. Goldthorpe, J. and generational Occupation Mobility in 1983. 'Inter- Northern Ireland and the Republic of Portocarero, L. generational Social Mobility and the Ireland'.BJS 38(1). Heath, A. 1981.SocialMobility.London: ConvergenceThesis'. BJS 34. Evans, M. 1987. 'Language Skills, Fontana. Language Usage and Opportunity'. Hope, K. 1975.Trendsin the Openness of British Society in the Present 21(2). Sociology Fiddler 1981. TheBritishBusinessElite. Century,Paper circulatedat the SSRC International Seminar on Social London:Routledgeand Kegan Paul. of Mobility:AberdeenUniversity. Giddens, A. 1973. TheClassStructure the Advanced Societies. London: Hope, K. 1982. 'Vertical and NonVertical Mobility'.ASR47(1). Hutchinson. Gilbert, N. 1986.'OccupationalClasses Hope, K. 1984. As OthersSee Us. London:CUP. and Inter-classmobility'. BJS 37(3). This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions 492 G. Payne Hornby-Smith, M. and Dale, A. 1988. Class Relationsin Britain'.BJS 23(4). 'The Assimilationof Irish Immigrants The Observer 1989. 'Maggie's Little Helpers'.Observer Magazine19.2.89. in England'.BJS 39(4). Hutson, J. 1987. 'Fathers and Sons: Parkin, F. 1979. Marxismand Class London:Tavistock. Family Farms, Family Businessesand Theory. in the FarmingIndustry'.Sociology 21(2). Payne, G. 1987a.MobilityandChange Society.London:Macmillan. Kaelble, H. 1981. HistoricalResearch on Modern Social Mobility. (Trans. F. Noakes) Payne, G. 1987b. Employmentand Opportunity. London,Macmillan. London:CroomHelm. Kaelble, H. 1986. SocialMobilityin the Payne, G. 1987c. 'De-industrialisation 19thand20thCenturies. LeamingtonSpa: and OccupationalMobility'. BJS 1987 38(2). Berg. Karabel, J. and Halsey, A. H. (eds) Payne, G. et al. 'A Re-appraisalof 1977.PowerandIdeology in Education. New "Social Mobility in Britain"'.Sociology 11(2). York:OxfordUniversityPress. Kelsall, R. K. 1974.'Recruitmentto the Payne, G. et al. 'OccupationalChange Higher Civil Service',in P. Stanworth and Social Mobility in Scotland Since and A. Giddens (eds) 1974. Elitesand the First World War' in M. Gaskin (ed.) The PoliticalEconomy of Tolerable Powerin BritishSociety.London:CUP. London:CroomHelm. Kelsall, K. et al. 1984Stratification (2nd Survivotl, Prandy, K. Stewart, A. and edition). London:Longman. Lee, G. 1981. Who Gets to the Top? Blackburn, R. 1982. WhiteCollarWork, London:Macmillan. Aldershot:Gower. Lipsett, S. M. and Bendix, R. 1959. Psacharopoulos, G. 1977. 'Family education and Social Mobility in IndustrialSociety. background, Berkeley:Universityof CaliforniaPress. achievement'.BJS 28. Little, A. and Westergaard,J. 1964. Raffe, D. 1979. 'The "Attendance 13. 'The Trends of Class Differentialsin Route"Reconsidered'.Sociology of Schooland Educational Opportunity in England Reid, I. 1986. TheSociology Education. London:Fontana. and Wales'. BJS 15. Lovatt, D. and Ham, B. 1984. 'The Richardson, C. 1977. Contemporary distributional implications of SocialMobility.London:FrancesPinter. Runciman, G. 1966.RelativeDeprivation de-industrialization'. BJS 35(4). Mach, B. and Wesolowksi, W. 1986. andSocialJustice.London:Routledgeand Social Mobility and Social Structure.Kegan Paul. Savage, M. 1988. 'Geographical London:Routledgeand Kegan Paul. Mack, R. W. et al. 1957.SocialMobility: Changesin Employment'.BJS 39(4). Thirty Years of Researchand Theory. Sorokin, P. 1927. SocialMobility.New York Harper. Syracuse:SyracuseUniversityPress. Marshall, G. et al. 1988.SocialClassin Stewart, A. Prandy, K. and Blackburn, R. 1980. SocialStratiRation Modern Britain.London:Hutchinson. London:CUP. Martin, J. and Roberts, C. 1984. andOccupations. Women and Employment:a lifetime THES 1989a.The ParentTrap. THES, 27 January. perspective. London:FIMSO. Miller, S. M. 1960.'ComparativeSocial THES 1989b. 'Would-be Students reveal biologicaldifferencesby choice'. Mobility'.Current Sociology 9(1). Muller, W. et al. 1988.'Educationand THES,2 February. Class Mobility'. CASMIN Working Westergaard,J. and Resler, H. 1975. Paper No.14, read at CASMIN Class in a CapitalistSociety.London: Heinemann. Conference,Regensburg. Noble, T. 1972. 'Social Mobility and This content downloaded from 188.254.76.113 on Sat, 20 Sep 2014 08:54:28 AM All use subject to JSTOR Terms and Conditions