ROAPE Publications Ltd.

The 'Overdeveloped' Post Colonial State: A Re-Evaluation

Author(s): Colin Leys
Source: Review of African Political Economy, No. 5 (Jan. - Apr., 1976), pp. 39-48
Published by: Taylor & Francis, Ltd.
Stable URL: http://www.jstor.org/stable/3997807 .
Accessed: 15/02/2015 05:07
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].

Taylor & Francis, Ltd. and ROAPE Publications Ltd. are collaborating with JSTOR to digitize, preserve and
extend access to Review of African Political Economy.

http://www.jstor.org

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

39

The 'Overdeveloped' Post Colonial

State: A Re-evaluation

Colin Leys
In this article Leysquestionssome of the formulationson the postcolonial state and its bureaucracydevelopedby HamzaAlavi, Roger
Murrayand John Saul. The articlesreferredto are Alavi's'The
State in Post-ColonialSocieties' (New Left Review, 74, July/August
1972) and Murray's'Second Thoughtson Ghana'(New Left Review,
42, March/April1967). But Leys is particularlyconcernedto open a
debate on John Saul's'The State in Post-ColonialSocietiesTanzania',publishedin TheSocialist Register(London, 1974).
At the end of his reviewof recent theorisingabout the state in postcolonial societies and its applicationto TanzaniaJohn Saul raisesa
criticalquestion:is state power in Tanzaniaa force which defends and
promotes the interestsof workersand peasants,or should 'the independent political organisationof progressiveelements, alreadya
(difficult) priorityin most other one-partyand militaryadministrative
regimesin Africa,become a priorityfor Tanzaniaas well'? (p. 367).
John Saul does not presumeto answerthis question;his concernis to
see whethercurrenttheory furnishesa valid frameworkwithin which
Tanzaniansthemselvescan try to answerit.
He startsout from HamzaAlavi'sinfluentialarticleon the state in
post colonial society, focussed on Pakistanand Bangladesh.Alavi
arguedthat (1) the originalbase of the state apparatusinherited-bya
'post-colonialsociety' lay in the metropole(ie it representedclass
forces existing there);its task was to subordinateall the indigenous
classesin the colony (ie it did not rest on the supportof any of them);
and hence it was 'over-developed'in relationto the ex-colonial society
(ie once it rested on the support of at least one indigenousclass after
the colonialistswithdrew).Specitically,it inheritsa strong militaryadministrativeapparatus.(2) The state directly appropriatesa large
part of the economic surplusand deploys it in bureaucraticallydirected
'development'activity. The 'centrality'of the post-colonialstate, which

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

evidently follows from these propositions,implies the 'centrality'of

the state bureaucracy.Alavisuggestedthat this bureaucracy,which he
called an 'oligarchy',was relativelyindependentof control by any
social class,but Saul points out that Alavi'sreasonsfor saying this are
not very clear;in any case, Saul notes, in East Africathere were no
strongindigenousclassesto be subordinated,so that the 'overdeveloped'
natureof the state is not due to the need to subordinatesuch classes,
but to the need to 'subordinatepre-capitalistsocial formationsto the
imperativesof colonial capitalism'.The absenceof strongindigenous
classesmust, however,affect the degreeof independenceof the state
bureaucracy;but just how, dependson why it was supposedto be
relativelyautonomousin the post-colonialsituation. If its relative
independencewas due to a 'balanceof power' between indigenousand
external class forces, then the absenceof stronginternalclasseswould
seem likely to make the state bureaucracysubservientto strong
external classes-ie the metropolitanbourgeoisie.On the other hand, if
the state bureaucracyis strong for other reasons,the absence of strong
indigenousclasses,as in Tanzania,might make it even more powerful.
Eitherway, Saul concludes,the state bureaucracyremainsvery
'central'.How should we seek to understandits significance?
He suggeststhat we have only two seriousalternatives:(a) the state
appropriatbureaucracyis a class of a new type, or a class-in-the-making,
ing and controllingproductiveresources,whetherby using state power
to acquireprivatecapitalor directly, in its capacityas the collective
'managers'of the so-called'public sector' (variousversionsof this
approachare associatedwith Fanon, Debray,and Meillassoux);or
(b) the state bureaucracyshould be seen as a fluid, still 'plastic'
category,largely'petty-bourgeois'in originbut, by virtue of its role
at the junction between the local economy and foreigncapital,exposed
to contradictoryinfluences-national, workingclass and peasantclass
interestsconfrontingthose of foreignand domestic capitalin the mode
of operationof the state. On this view they are a potential new class,
but not a class whose charactercan be known in advance,but only
from the emergingresultsof the way they grapplewith the contradictions in their conduct of affairs.(This conception was first
formulatedby RogerMurraywith regardto Ghana).
Turning,finally, to Tanzania,Saul arguesthat only the second of these
approachescan really accommodatethe progressiveinitiativesof
PresidentNyerereand his supportersfrom 1967 onwards.The former
approach,which Issa Shivjiadopts in his analysisof the class struggle
in Tanzania,involvestreatingall these initiatives(from the Arusha
Declarationonwards)as essentiallydeceptions, exercisesin ideological
manipulationin the narrowclass interest of what Shivjicalls the
'bureaucraticbourgeoisie',or at most as reluctantconcessionsto
popularforces. Saul does not underratethe pressureswhich operate
to curtailand neutralisethe influence of the progressiveforces inside
the state bureaucracybut he contends that their initiativeshave been
genuine, and that it is essentialto understandthe 'plasticity' of the
state bureaucracyin terms of a real classstruggletaking place between
different elements inside it. The question confrontingTanzaniais then,
in these terms, whetheror not the bourgeoisforces have effectively
won in that arena,necessitatingthe 'independentorganisationof

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

progressiveelements' alreadyreferredto.
The OverdevelopedState?
It is not difficult to agreethat the ArushaDeclaration,ujamaavijijini,
Mwongozo,etc., have resultedfrom some kind of struggleswithin the
state apparatus.But Saul himself is now pessimisticabout the prospects of furthersuch initiativesfrom that quarter,and there is mounting primafacie evidence of the weaknessand/or neutralisationin
practiceof severalof those which were taken earlier;in a recent
sketch Aidan Foster-Carterrecently went so far as to argue,with uncomfortableplausibility,that the portrayalof Tanzaniaas a country
makingthe 'transitionto socialism'is and alwayshas been a myth, and
that the realityis one of the last of an old line of 'populist' regimes,
stretchingfrom SukarnothroughNkrumah,and one whose days are
also numbered.But what, then, is the practicaldifferencebetween
John Saul'sposition and Shivji's?Saul'sseems to consist in leavingit
to 'those engagedin significantpraxiswithin Tanzania'to determine
whetherthe strugglewithin the state bureaucracyis really over, whereas
Shivjiholds that it hardly,if ever, really occurred.In fact it is not
clearthat Saul'stheoreticaldiscussionof the state reallyilluminates
the key question he is posing;to my mind it is more illuminatedby
his variousreferencesto other issues,such as the natureof TANU, the
characterof the industrializationpolicy, the mobilisation/demobilisation of peasantpolitical action, etc., many of which he has discussed
quite fully in previousarticles.The reasonfor this, I suspect, is that
the theoreticalformulationsabout the state which he has surveyedin
this article are defective.
Let us begin with Alavi'sconcept of the 'overdeveloped'state inherited
from colonialism.Whatdoes 'overdeveloped'really mean here? The
word suggeststhat the inheritedstate apparatusis larger,its coercive
or administrativepowersweightieror more ramified,than they would
be if the colonial state had not had to subordinateall the domestic
classesincludingthose which were themselvesdominantclassesin the
pre-colonialsocial formation.By dubbingthe colonial state a 'powerful bureaucratic-military
apparatus'Alavireinforcedthe superficial
plausibilityof this; the 'overdevelopment'of such states then became,
for him, an accepted fact which his class analysisof its historical
originscould then explain. But a brief reflectionsuggeststhat this is
misleading.Evenif it were true that the colonial state apparatuswas
more powerfulmilitarilyand administrativelythan it would have
needed to be, if it had not had the task of subduingnative kings and
princesand their rulingclasses,this does not mean the force at its
disposalwould necessarilybe excessive for the tasks of dominationin
the situationwhich existed by the time formalindependencewas
achieved.For by that time the capitalistmode of productionhad been
introduced,and made effectively dominant,in the colonial social
formation,givingrise to a new and developingstructureof class
antagonisms.In fact it seems more plausibleto arguethat the colonial
state, after the initial resistanceto conquest had been overcome,
disposedof less militaryforce than it would have requiredif it had
not been able to rely on reinforcementsfrom the metropole or other
partsof the colonial empirewheneverthe need arose. At any rate,
in the Indiansub-continent,which Alavihad primarilyin mind, the

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

41

42

civil serviceand armedforces were expandedmore rapidlythan either

nationalincome or populationin the yearsfollowing 1947, and in fact,
everywherethe expansionof the post-colonialstate seems to have
been dramatic.Of coursethis does not prove that it was not 'overdeveloped'alreadyat independence,but it certainlyis true that the
states of the ex-colonies actually tend to be small, relativeto both
populationand the size of the economy, comparedwith the states of
the advancedcapitalistcountries.The relativelylow shareof national
income taken by governmentrevenueand expenditurein underdeveloped
countrieswas noted ten years ago by analystsof the bourgeois'developmental'school, such as Russett. For our purposes,a few caseswill
illustratethe point:

Tanzania
Kenya
Uganda
U.K.
W. Germany
U.S.A.

3.

1.

2.

National
income

%GDP from Central govt.

agriculture budget

37
?431m.
31
?563m.
49
?475m.
?46,OOOm. 3
DM611,000m. 3
3
$879,000m.

4.
(3) as %
of (1)

25
?107m.
20
?115m.
16
?80m.
37
?17,525m.
DM148,000m.* 24*
42*
$370,000m.*

(Budget figures marked * are for Federal and State/Lander governments combined. All data from UN Statistical Yearbook, relating to the year 1970.
Because of differences in the way public expenditure figures are defined they
are only roughly comparable between countries. Total 'public sector' expenditure in Britain in 1970 was 48% of G.D.P.)

Of course the reasonwhy the shareof nationalincome appropriated

by the state is smallerin underdeveloped,or periphery-capitalist
countries,is that the volume of surplusvalue producedthere is small;
so it may well be that in such societies, the state directly appropriates
a largershareof the surplusthan it does in countrieswhere much
more is produced,althoughwe cannot assumethis without further
investigation.Certainly,the more of the total volume of surplusvalue
which the state directly appropriates,the more it is likely to be
immediatelyinvolvedin the class struggle(by contrastwith situations where the state merelyguaranteesthe conditions for the
appropriationof surplusvalue by the bourgeoisiethroughthe
apparently'natural'operationof the labourmarket).All this, however, is not describedor illuminatedby callingsuch a state 'overdeveloped'.
In the case of Tanzania,Saul himself notes that the colonial state did
not encounterany strongindigenousclassesto be subdued(this was
true throughoutmuch of subsaharanAfrica),but he still maintains
that the Tanzanianstate was 'overdeveloped',sayingthat this was due
to the need to 'subordinatepre-capitalistsocial formationsto the
imperativesof colonial capitalism'.But why should this call for a
particularly strong state if there were no strong classes to defend their

interestsin the old social formations?In any case, it seems to me

simply implausibleto regardthe colonial Tanganyikanstate as very
strongin relationto its tasks, let alone as 'overdeveloped'.John Saul
continues to subscribeto this idea, I suspect,becauseit suggestsa
reasonfor the 'centrality'of the state in post-colonialsocieties. But

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

surely 'centrality'is anotherempty word in this context? The state

is equally importantin all class societies; it is no more 'central'in
Tanzaniathan in Britainor the USA (or the USSR). It may be more
'embracing'(ie may own more productiveforces or intervenemore
directly in variousareasof social life) in some societies than others,
but in this respectit is typically less 'central'('extensive' would be
a better word here) in most post-colonialsocieties than in most
advancedcapitalistsocieties.
But the importantpoint is not so much that the idea of the 'overdeveloped'state is empty; it is really that this whole way of
approachingthe question of the significanceof any state, ie of starting out from its structureor scope, whether inheritedfrom an earlier
situation or not, is a mistake.In orderto understandthe significance
of any state for the classstrugglewe must start out from the class
struggle,not the state. The idea of the 'overdevelopedstate' functions,
in both Alavi'sand Saul'saccounts, as an apparentreason for reversing the properorderof procedure:the inheritedstate is said to be
'overdeveloped',thereforeit has exceptional significancein postcolonial societies, therefore.theclass characterof the bureaucracyof
this overdevelopedstate is the key issue. This leads to formulations
about the state bureaucracywhich seem as questionableas the idea
of the 'overdeveloped'state.
First, the discussionblurs importantdistinctionswhich need to be
made between different elements in, and branchesof, the 'state
bureaucracy'.Second, the 'class character'of the state, ie the class
interestswhich state power reflects and promotes, is confused with
the class membership,or class position, of the functionariesof the
state apparatus.Third,both sides in the debate about the state bureaucracy use a ratherloose concept of the 'petty-bourgeoisie'which they
regardas the class from which the state bureaucracyoriginates,and
whose interestsin reflects;this is true whether they think this means
that it tries to use its position in the state to turn itself into a
bourgeoisieproper(as Shivjiholds) or whether(like Murrayand Saul)
they think this means that the use they will make of state power is
initially indeterminate.Let us briefly considereach of these points in
turn.
(a) The 'state bureaucracy'
Withinthe state apparatuswe need to distinguishbetween two
elements, even in 'party states' of the Tanzaniantype. First, the career
bureaucrats,and second, the personnelinsertedinto, or added onto,
the apparatusthroughthe 'statification'of the nationalistparty.
Careerbureaucratsinitially include, especiallyin senior posts, a high
proportionwho were recruitedundercolonialism,but they arejoined
by youngerofficials whose formaltraining,entry route and career
aspirationsdo not differ very much from theirs. In spite of some interchangebetween state and 'party' posts, especiallynear the top of the
hierarchy,the two elements remaindifferent in one importantrespect,
at least for some time: the party-recruitedelement bringswith it into
the state apparatuspolitical links with workersand peasants.Murray's
argumentprimarilyconcernsthe party (CPP)leadership,ratherthan
the civil service;Shivji,on the other hand, is particularlyconcerned

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

43

44

with the expandingnumbersof managersof state-ownedenterprises,

the 'bureaucraticbourgeoisie'par excellence. Saul notes this difference
but does not stressit, and it is true that in Tanzaniathe policy of
'cross-appointing'party leadersand careercivil servantsmakesthe
formaldistinctionseem less important,yet I think it may be significant for understandingthe originsand course of the strugglesthat do
occur inside the state apparatus.It may also be importantto distinguish between both levels and branchesof the apparatus.Saul'sview
involvesidentifyingprogressiveelements at policy-makinglevels
whose initiativesare severelyemasculatedby the 'dead hand' of 'the
bureaucracy'(often lower level officials chargedwith implementation).
Thisis a differentprocess from that of the party leaders'crystallising
as a privilegedclass aroundthe apparatusof the state' (Saul in RAPE
No. 1) which may of course also be true for other leaders.The
distinctionbetween branchesof the state may be importantas the
state takes over more and more of the economy. The officials entering the 'state enterprise'branchare especiallyexposed to the
bourgeoisvaluesembodiedin the technology, managementpractices,
'efficiency' ideology, etc. of the firmsthey take over, especiallyas
they are dependenton the formermanagementfor advice and are
thrustinto a ready-madeclass position vis a vis the workers.
(b) The 'class character'of the state bureaucracy
One of the valuablepoints made by Nicos Poulantzasin his book
PoliticalPower and Social Classesis that it is a mistaketo think that
the class origins,class ties or class ambitionsof the individualswho
compose the apparatusof the state need be the same as those of the
dominantclass, or that of state power reflects their own class interests,
except in a secondaryway. The first question must always be which
classis dominantin a given social formation,since this dominance
must be enforced by the state; the class characterof the state is given
by this relationship.Of course the relationshipis not static. But the
class interestsof the state bureaucracy,whetherthey are congruent
with those of the dominantclass, or in conflict with them, are unlikely to be the determiningfactor in establishingor upsetting that
dominance.Or to put it anotherway, even if the state bureaucracy
enjoy great'relativeautonomy', and have a distinct class interest of
their own, it doesn't follow that the class characterof the state, or of
state power, reflects this interest. This seems to me true even in the
USSR, from which this whole problematicprobablyderives.Trotsky,
who first posed the question whetherthe state and party bureaucracy
had developedinto a new class, clung for a long time to the Leninist
idea that it was acting as the agent of the new dominant class, the
proletariat;when he finally came to abandonthis belief, he was forced
to postulate that some other class ruled;and since this was clearlynot
the formeraristocracy,or the formerbourgeoisie,he concludedthat it
would have to be the state bureaucracyitself. But the originalityof
the Russiansituation permitsother formulations.If it was not a classless society, it did not necessarilyhave a new dominantclass;those
who have describedit as 'bureaucraticcollectivism'have in partbeen
makingthis point. At all events, in post-colonialsocieties in Africa
there can be little doubt that the dominantclass is still the foreign
bourgeoisie;and the question then is whetheror not the Tanzanian
state servesits interests,not whetherthe Tanzanianstate bureaucracy

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

has 'become a bureaucraticbourgeoisie'by virtueof its control over

state-ownedcapital.To repeat;we cannot discoverthe class character
of the state by inspectingthe class interestsof the state bureaucracy,
even if these are clear;Murray'ssuggestionthat the 'political class'is
subjectto 'multiple determinations',and that its classsignificance
must be discoveredin its 'modus operandi'of state power confuses
the issue, not only becausethese phrasessay so little, but becauseit
points our attention in the wrongdirection.
(c) The 'petty-bourgeoisie'
In the debate about whetherthe state bureaucracyare a new class, or
a class-in-the-making,
etc., it is generallytaken for grantedthat the
membersof the state bureaucracyare drawnfrom the 'petty bourgeoisie'. If this debate is misconceivedanyway, it may seem unimportantwhetheror not their class originsare being correctlydescribed.However,the rathercasualway in which the term 'petty
bourgeois'is used correspondsto the lack of interestwhich some of
these theoristshave displayed(at least in their discussionof the state)
in what I would call the historicaltendenciesof the capitalistmode of
productionin these 'post-colonial'societies, a question which seems
to me fundamentalfor anyone attemptingto answerthe political
question posed by Saul-ie how we should assessthe significanceof a
regimesuch as Tanzania'sin 1975.
Briefly, Marx'suse of the term 'petty-bourgeoisie'was historically
fairly specific. It referredto small manufacturers,shopkeepers,peasants and artisans:
In countries where modern civilisation has become fully developed, a new class
of petty bourgeois have been formed, fluctuating between proletariat and
bourgeoisie and ever renewing itself as a supplementary part of bourgeois
society. (Communist Manifesto).

It was a 'petit' bourgeoisclass in the sense of being in possessionof

smallamountsof capital,and hence havingan interestin the preservation of privateproperty,and hence havingan interest in the presevation of the power of the bourgeoisieproper.On the other hand, its
interestswere also opposed to those of the bourgeoisie;individual
membersof this class,however,arebeing constantly
hurled down into the proletariat by the action of competition, and, as modern
industry develops, they even see the moment approaching when they will
completely disappear as an independent section of modern society, to be
replaced, in manufacturers, agriculture and commerce, by overlookers,
bailiffs and shopmen. (Ibid.)

The political consciousnessof the petty bourgeoisreflected this

ambivalence:
. . . he is dazed by the magnificance of the big bourgeoisie and has sympathy
for the sufferings of the people. He is at once both bourgeois and a man of the
people. . . He is himself nothing but social contradiction in action. (Marx to
Annenkov, 1846.)

It is a 'transitionclass, in which the interestsof two classes(the

bourgeoisieand proletariat)are simultaneouslymutuallyblunted. ..'
(EighteenthBrumaire)

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

46

By contrastthe word 'petty bourgeois'as used by the partiesto the

debate on the state bureaucracyin post-colonialAfricarefersmainly
to (i) ownersof small amountsof non-agriculturalcapital,such as
small manufacturers,contractors,traders,etc. (ii) the richerpeasants (iii) white collarworkersgenerally(mainly, of course, in state
employment). This clearlymeanssomethingdifferent from Marx's
concept. For one thing, the last categoryare not owners of capital
at all, but the sort of people who Marxthought would replacethe
petty-bourgeoisieas he used the term, even though they may well
have 'bourgeois'tastes, ideas and aspirations.The richerpeasantsare
generallyincluded, and other peasantsexcluded, because the former
have distinct economic intereststo defend againstthe latter and
againstthe rurallabourers.Further,all these categoriesare typically
expandingin present-dayAfrica,and have gainedboth influence and
wealth at the expense of the rest of the peasantryand the majority
of wage workers.
The 'petty-bourgeoisie'so defined is thus a different concept from
that of Marx.This does not necessarilymean it is inept, but it does
mean that its political implicationscannot be taken for granted.
WhereasMarx'spetty bourgeoisieplayed an ambivalentpolitical role
correspondingto its contradictoryclassinter'estsvis a vis the developing bourgeoisieand proletariat,this seems less likely to be true of the
'petty bourgeoisie'as the term is used by both Murrayand Shivji.In
orderto know the real significanceof any statement about the 'petty
bourgeoisie'as they use it, we need a generalanalysisof the development of the capitalistmode of productionand its relationswith
petty commodity production,and thus of the developmentof the
relationsof productionand the class struggle.And this is also the
essentialstartingpoint for an assessmentof the role of the state-and
of the significanceof the fact-if it is significant- that its personnel
are recruitedfrom one class ratherthan another.
Conclusion
The point of these commentshas not been to disagreewith Saul's
interpretationof the situation in Tanzania,but to question whether
the theory he has used in this instance reallyhelps to illuminatethat
interpretation,or points the political way forward.
As far as the analysisof events in Tanzaniais concerned,the distinctions I have tried to drawat most put a gloss on Saul'sviews, especially
with regardto the originsand the course of the strugglesinside the
state apparatus.The originsof any such struggleevidently lie in the
links-personal, organisationaland ideological-with the workersand
peasantswhich some of the party-recruitedelements bringinto the
state apparatuswith them. Individualcareerofficials identify themselves with their outlook, but the impetus comes from the party
elements, and it was in the party executive, by then largely composed
of holders of state posts but still organisationallyand ideologically
distinct from the state, that the initiativesof 1967 and 1971 were
taken. These initiativesreflected an appreciationof some of the class
implicationsof the existing social and economic system, as revealedin
a successionof policy contradictions(neutralismversusdependence
on bilateralaid, egalitarianismversusthe elitist educationalsystem,

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

etc.). This appreciationwas very partial,however. In particularit was

assumedthat the dominanceof a local bourgeoisclass, and of foreign
bourgeoisies,could be preventedby legislativeand administrative
action taken by the existing state. This ignoredboth the bourgeois
characterof the existing state (its adaptationto the task of defending
bourgeoisinterests)and the fact that the penetrationof Tanzanian
society in all its dimensionsby capitalismwas far too advancedto be
checked, let alone prevented,by juridicalmeasures.Rosa Luxemburg's
words apply as much to peripherycapitalismas to capitalismin the
metropoles:'. . . the fundamentalrelationsof the dominationof the
capitalistclass cannot be transformedby meansof legislativereforms,
on the basis of capitalistsociety, because these relationshave not been
introducedby bourgeoislaws' (italics added). To check, let alone
eliminate,the dominanceof the capitalistclass could only mean
mobilisingthe workingclass and the poorer peasantsto struggleagainst
it at all levels. This was excluded, partly by Nyerere'sresistanceto the
idea that class strugglewas involvedin the 'buildingof socialism',and
partly, one suspects, from a reasonablefear that massstrugglewould
involve the leadershipin being outflankedon the 'left' by new leaders
emergingin such struggles,while simultaneouslyrunningthe risk of a
reactionfrom the rightwithin the state apparatus.Distortion and
'neutralisation'of the initiativestaken by 'Nyerereand his supporters'
within the state apparatusclearlyplayed a partin reducingtheir
impact-the clearestexamplebeing the open opposition to the urban
workers'responseto the TANU Guidelinesby the economic bureaucracy-ie state-companymanagersand directors,supportedby the
union bureaucracyand the police-in 1973-74. But it needs to be
emphasisedthat this processitself reflects the limitationsof the
originalinitiativestaken. It was Nyererewho insisted in February
1967 that the nationalisationmeasureswere 'primarilynationalist'
in purpose,and that the fact that ownershipand control had to be
transferredto the state, which made the measures'socialist',was
welcome (because 'we are socialistsas well as nationalists')but
incidental;in the interestsof 'Tanzania'the firms affected 'shouldbe
efficiently run. Theirmanagementmust be good, and their workers
must play a full part in securinghigh production.Industrialdiscipline
is an essentialpart of this process. . .' (Economic Nationalism).These
ideas markthe limits, not of one man'sideology, but of the broadly
'populist'form of consciousnessof the wing of the originalnationalist
leadershipmost sensitiveto its massbase. The 1973 decision to enforce 'villagisation'need not necessarilybe seen as completely inconsistentwith Nyerere'searlierpolitical position, either. His 1968
statementwas that no-one could be forced into an ujamaavillage;it
could not be a socialist villageif force was used. The villagesinto
which people have since been forced are actuallytermed 'development
villages'.Whilethe Presidentwas obviouslyvery reluctantto use force,
for any purpose,he seems to have concludedthat it was justified in
orderto improvethe ability of the governmentto direct an increase
in agriculturalproduction.This is consistent with the conception of
'socialism'as somethingthat can (if necessity requires)to 'introduced'
later, when the materialbasis for it has been built.
Lookingahead,the theoreticalconsiderationsgiven earliersimply
suggestthat the concept of the 'bureaucraticbourgeoisie'as an

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions

47

48

entrenchedrulingclassis an unreliablestartingpoint for further

analysis.Most writerson Tanzaniaare impressedby the state bureaucracy'sexpansionand growingpowers, and by the bureaucratisation
of the party and the trend towardsauthoritarianismin general.But
the contradictionsof the situationare obscuredby this lumping
together of differentelements in the state apparatuswith the idea of
the dominantclass and specificallyalso with the undifferentiated
'petty bourgeoisie'.One illustrationmust suffice to indicate the sort
of issue involved.The initiativesof the ArushaDeclaration,etc.,
cannot be known to be 'progressive'per se, but only from an analysis
of the class forces and contradictionsof the situation as a whole. So
the 'progressiveness'of the nationalisationmeasurescan be questioned: perhaps,as Aidan Foster-Cartersuggests,they were reallyjust
'the most up-to-dateform of denationalisation'.Conversely,can the
enforced villagisationbe known automaticallyto be retrogressive?As
Raikespoints out, it does not obviously advanceany class interest of
the state bureaucracyand it could both raisepeasantconsciousness,
and even stimulateorganisationamongthem. Or again,how well does
the presentpower of the state bureaucracycater to the materialneeds
of the non-statepetty bourgeoisie-rich peasant(included in forced
villagisation?),traders,entrepreneursof variouskinds?These
questions seem prejudgedby the propositionthat the state bureaucracy has now constituted itself a rulingclass.
In general,I am not entirely convincedthat the 'state bureaucracy'does
now constitute a class, ratherthan havingsimply consolidateditself in
its bureaucraticfunction, in howeverofficious a manner,and with however bourgeoisa mentality;or that the bourgeoisieproper(abroad)has
been really, ratherthanjuridically,expropriated,underthe nationalisation arrangementsas these have been describedby Shivjiand others;
in short, I am not sure that there has been a fundamentalevolution in
the relationsbetween the differentelements of the rulingclass alliance,
as the 'rulingbureaucraticbourgeoisie'thesis implies. Unless these
issues are first clarified,the next phase of the class strugglecannot be
clearlyunderstoodeither;specifically,it is not a question of deciding
whetherthe strugglebetween the 'progressive''nizers'and their
opponents inside the state bureaucracyis over, but of reconsidering
all the changesthat have been made in terms of their impact on the
developmentof the class struggleas a whole, and what that impact
now implies for future strategy.

This content downloaded from 14.139.196.3 on Sun, 15 Feb 2015 05:07:15 AM

All use subject to JSTOR Terms and Conditions