PERSPECTIVE

Roger leakey

21534

A Njenga

A groforestry has been definedin several ways (Nair 1989).ICRAF's current definition is acollective name for land-use systemsand practices in which woody peren-nials are deliberately integrated withcrops and/or animals on the sameland-management unit. The integra-tion can be either in a spatial mixtureor in temporal sequence. There arenormally both ecological and eco-nomic interactions between the

woody and non-woody componentsin agroforestry. This definition has

Roger Leakey is Director ofResearch at ICRAF, PO Box30677, Nairobi, Kenya; email:rleakey@cgnet.com; fax: +254 2521 007.

january - March 1996

served well and helped agroforestryto become recognized as a branch ofagricultural science that is rapidlybecoming a science in its own right(Sanchez 1995).

Agroforestry practices come in-many forms but fall into 2 groups-those that are sequential, such as fal-lows, and those that are simultane-ous, such as alley cropping (Cooperand others, in press). In all,some 18different agroforestry practices havebeen recognized by Nair (1993), al-though each has an infinite numberof variations. Thus, at the moment,agroforestry is viewed as a set ofstand-alone technologies that togeth-er form various land-use systems inwhich trees are sequentially orsimultaneously integrated with crops

and/or Iivestock. In agroforestry re-search, practices are often appl iedafter diagnosis and design, participa-tory research or characterizationstudies, as appropriate, dependingon the social, economic and envi-ronmental problems in an area.Agroforestry is generally practisedwith the intention of developing amore sustainable form of land use

that can improve farm productivityand the welfare of the rural

community.My problem with the current view

of agroforestry is that many peoplestill see it as a set of distinct prescrip-tions for land use. As a result, it fallsfar short of its ultimate potential as away to mitigate deforestation and landdepletion and thus alleviate poverty. A

5

PERSPECTIVE

different view, however, is that agro-forestry practices can be seen as phas-es in the development of a productiveagroecosystem, akin to the normal dy-namics of natural ecosystems. Overtime, the increasing integration oftrees into land-use systemsthroughagroforestry can be seen as the pas-sage towards a mature agroforest ofincreasing ecological integrity. By thesame token, with increasing scale, theintegration of various agroforestrypractices into the landscape is like theformation of a complex mosaic ofpatches in an ecosystem, each ofwhich is composed of many niches.These niches are occupied by differentorganisms, making the system ecologi-cally stable and biologically diverse.Filling some of these niches withspecies that provide important

environmental services or economi-

cally valuable products or both shouldresult in land use that is both sustain-

able and productive. Furthermore, thebenefits with increasing scale from thefarm to the landscape and the regionare exponential, since the ecologicaland social benefits of diversity on alands0:apescale are considerablygreater than the sum of the individualfarm-scale benefits.

Within this ecological framework,farmers can manipulate and managetheir trees to take advantage of thebenefits of the processes in ecosystemservices or products, by breaking theprocess of agradation-or ecosystemdevelopment-at any point, or by al-lowing a mature agroforest to devel-op. Fallowing and relay cropping, forexample, make use of the benefits of

the early successional stages ofecosystem development, while com-plex multistrata systems approach amature vegetation, such as the com-mercially valuable damar agroforestsof Sumatra (see article this issue p 8).

Therefore, I suggest that agrofores-try should be reconsidered as a dy-namic, ecologically based, naturalresource management system that,through the integration of trees infarm- and rangeland, diversifies andsustains smallholder production forincreased social, economic and en-vironmental benefits.

If the above concepts are accepted,then agroforestry researchers andextension workers have a new chal-

lenge-to start the process of inte-grating a number of the currentagroforestry practices into productive

Jamie Wyant

Agroforestry - an ecological perspectiveabout the desirability Qtion.

Managing th cological integrityof an agrof estry system dependson an' erplay between the socialva s within the system and its

cological potential. This requiresan understanding of the natural cy-cles of an ecosystem-of a system'scapacity to withstand human andnatural disturbances and still main-

tain its 'preferred' condition-defined by human values.

Like the natural recovery processof disturbed forest ecosystems, theintegration of trees into agriculturalsystems through agroforestry can beseen as the passagetoward anecosystem of increasing ecologicalintegrity (figure 1). After it isdisturbed, a healthy ecosystem maynot return to a pre-disturbance statebut it can re-establish its normal

range of dynamics. So too with anagroforest-there is no single 'target'for the ecological integrity of such an

sure and manage the ecological in-tegrity of an agroforest. It is impossi-ble to define ecological integrity in apurely objective way because it alsorepresents the natural products andservices that humans may value inagroforestry ecosystems-and human

values are subjective. For exampl~areas where livestock is imp~ant,fodder is going to be highly/Valued,

but in the absence ofjive~iock, crop

residues ma:ypresel1fa disposal prob-lem. Or wome ay value trees

more for tf), medicines they produce,as thev~d to be responsible for thehe}1th of the children.

Thus we measure and analysechanges in an ecosystem but wecan only make judgements aboutthe integrity of the system. Ecologi-cal science can provide the neces-sary information to assessthecondition (or state) of an agrofores-try system, while social sciencesprovide the information necessaryto make value-based judgements

Jamie Wyant is a senior ec ogistwith ICRAF and diagn IC leaderfor the African Hig nds Initia-tive, PO Box 30 7, Nairobi,Kenya; email-jwyant@cgnet.com;fax: +254 521 001.

At the moment, agroforestrypractices are generallyviewed as different, staticland-use systems. Yet, current eco-logical theory views ecosystems asdynamic and self-organizing.Ecosystems, like the living organismsthey contain, are changing continu-ously, assuming a succession of'states'. The state of something is itscondition as described by a list of thevalues of its properties-any charac-teristic or attribute that can be evalu-

ated quantitatively. But ecological in-tegrity involves more than just eco-logical principles. Social values alsohelp shape how we understand, mea-

6 Agroforestry Today

I, " I' I ( r I V F

dlHI ,u,ldlridiJle lanel-use systems

Ihdl dll('Vldtl' poverty. Contrary to Ill('•dtl'lI1dtlvc of rnonocultures, over

lirm~ ,md spdce these land-use sys-

Il'nb iJu olne rnore complex, iJiodi-

V('fc,e ,lIId IJoth ecologically anel('( onol11ic,llly resilient to the normal

pdltell1c, (II (11f11dtic variahility anelpest ,md disease outbreaks. It IS

worth noting that in some areas 01

high popul,ltion density, farmers ,He

,drcaely ahead of the game, anel ,He

prdclising this kind of agroforestry. ~

References

Cooper PI/v\. Ledkev RRB, RJO MR ,lI1e1

I\CVllolcis L III press. Agroforestry .lIiel

the ll1itigdtioll of lallcl depletioll in IIH'

hUlllici dllcl suh-hull1icl tropics Of

'\lrICd h{!l'riI1lL'/JI,,1 Agriculture.

'-dlr I'KR I 'IWI'\"loiorestry clerilH'eI I'

I: III /11 1'''-I~ ,,",,III •(·ell

;·\f..;((J/(IF(· ....tn ",\ "{('Ill'- in the

I(,('{!I( '. i)(,lell