MA.NACI nc/teaching new analysts
Martin Felnwn
We haveon there,be en bate wW hat >Wtlake* dry-me. vooiuAeadfcoe* ii* eraaea mi eektln- and more ba mtf in Lol. thinkinc "Wm cant chinaa ti now."of every ibtnf. The prlnl burton on Kbaped down andscrccb-los meltdown Hi* in-hoi looki like lhe reaWry H* refutei todliectmra. and became ill hit prolr-aon told hhn he -rue*iutl be your faulltuttlmev
You're rtgha. Pum lima lint Buy it not enoueb. Uul what ilwut lb* new litre whoimooaJy tr>.ne, mm lo hive tbe latent,ill flracikns' Onewrf acn repnatndh if tneof oew analvata al Ihe end of theM EfKa daawetaryaa]Wow to douapect Truce rarateoiUilllfniH mimmrrt wmii tha -eirononrted analiiti in iWu pnelMmi penodercent ll tuny
JiT'Hi"
aniccreuedJ producttm. crovuluit mil hand kildtnt terntei aurtnt tbe adwtroenl peood. and paulm on tribal lore Ciowth hat tinisher percentage olmanatan ttrunltns with theae awm lor the But time
After attempt mt to copehile, many manairrt aie (sad*onclude that inalyiii are born,m nal rend) to abandon entirely btt one thai mayit ertreme Like the old jrAci however, tbe newt L> goodbad Tbe bod la thaioechanical formutaa that an ones* new bire an (onW to eitaraniee an acceptable piece oi batdaed MeKsencc Each analy* nruj) team the alb aa each before kin bscy Uyrac laJanaj abort, and tnical ant*e Bncro-wd n* the iKingt ia banned al the aneanent ofula* deration wit- what Mark Twain batd to be tbe neat important Ingiedkmt in naecrst "i'd rather be lucky thane aaai
S nawa tt that the manage: can facilitate the learning, process If he or ihe can
a irree ol our minion and trie diHnrent* between tnte*-liBente wnlirm andwrtttnai
- Pearnbe the vnera ol iieclhfrfice analytlilear, com* iathien
eaatyifca-lurea and prc.de tab of peaiHi* rrWcaxeaaenl and
The FVtl Step
na haoltiformed In (cur to ten ynan oi ernitee level work. ThUeomo down to impteMirat on the wpvU-be mailar o( what lhebarougb diarunion of the
naturenwflin!ocr wilting. There are seven key concept* that lheanatyti mml abveeb. ihree relating lo Iri. mission and lour ir> Intelligencem mn.lnced that Ihe branch chief must hammer (If need be) (he "threeand the four essences"into the new hires bead, or face the proapcel of yn another (ournevrnari who coninbuics little but can up enormoui amount! of managerial time.
The job it lo male judgment, about Ihc future. The new analystdifficulty accepting; the idea weoncerned about whatthan the tiem6eanee of lhe event for US .nteresi* Moreover,bv college to anarch forandto
than what i. actually the fact. And then there ts thatdetermination lo get all theompoteoo reinforced bv the mfetakcit notion that ouro know everything.
The manageret the new analysi to understand that-
JutJgroents wlfl invariablyhe baiit of incomplete and
Tr*rearerx>"iiicornpIetcs"alvenbfre Theaaalyat never ha* theluxury of asking the consumerittle more time for the situation to clarify or to wail until additional information become amiable.
i always sounds, our tab is noi so much lo be fight as it is to provide lhe best answer pcraiUe given the time and information available.'
We are the interpreted of foreign culture, and alien problems. At such, out X* it to etpoae the loefc behind ihe actionsiddle Etui nudntan and to render intelligible io the aeneral reader tbe phyocsoviet barrage attackS missile system
Our job is to support decision makers. Th*oncept thai all new analyst* readily accept. Indeed, for many, the piotpect of being part of the pobcv Piece* is one of tbe .Longest sellir* pant, cf the iob Pol (there i*oncept that many new analysis have difficulty putting into practsc* because they are contused about what constitutes support
Many believe that if ihey add to thenowledge, they have done the job. The ra.oarer muttthai the potrn of analysts is the inier-preutton cf Inforrnanon, not itt presentation. Analysts mutt be taught to giaw the distinction between providing answers to ical problem! and expanding lhe
Mmiagiig/Ttiuhiiig iVeiv Analysis
body of knowledge on some subiect Supporting ihr pork-maker comet down lo three related function*
Proitding answers to specific quotient, only tome ot wfuch may be asked b> the poiicvDukci.
Piculing tillumination ot the forces at work aodear onallow* tb* policymaker to understand aond lo proem* new information-
Where appropriate, to wain.
The (oar "essences" of IntelligenceBow directly fromut* dealing with Ihe mimon The table sumnnriies the differences between sca-derate and intelligence writing.
lh< Difference Between Academic and InteUiaence
Facia(hien tbe KTVBK
Wntwn if EXPFJtTI wxh iulor GENErULISTS sntftlint -ll>
ioiwfcUim
Dnalltd.uf*y. munantf^ chaiMiallaOHB
Shcittaonwith eunclinlsm aad cipken then
riting focuses on the future and it is not enough lo marshal
iheew analyst who was struggling to make the trarunticn from academe captured this problem best. In college ihegathered all the facts and "the conclusion* ps fetJtanv ate under th* mramrwessiort that the primaryot intelligence writing: it to discover truth Or set the record martin, andesult, the firsto ky out In (ovine detail how the pietent situation evolied.
Managers need to impress on new analysts that because what people behev* to be Hue Is often mere important than what if. discovery of the factsi jfieni for and occasionally immaterial to the real lob ofabout the future Students become analysts when they stop thinking in terms of what happened and start thinking in tenxs of what the facts mean Many just never seem to makf this transition
Intefrlgrnce It written for aeneralUti who are grappling with realOne ot theines for new inajysrs to grasp Is the nature of their audience. Unlike the situation lo which they ww Bnd themselves, tbey arc used to writing for professors and academics who demand detail and who are under no ceUgatio- to do anything with the information. New people are alto stow to realize and often doubtew months on ihe lob they are among the most knowledgeablehe governmentarticular Issue, and for tbe Brst time rn their Uvea, tbey are writing for an audience that
Martin Petersen
Managing/Teaching New Analysis
Tcoening
ompleltna tbe research, identifying lhe two or three key point* theoy from the paper.
tiling questions lo organiae (he pacer.
Slop one Identify the intelligence lasuc within an UiieltgeDoe topic. Or. deal -nth the "Cie* Title Trap"ew analyst what he is writing about and odds awwilleply aiooa the lines ol "corruption in China" or -defense strategy in" or "Soviet activity tn the Thodnd what you get i) everything about Soviet acnvitvhe Third World starting with Afghanistan and proceeding through Zimbabwe.
Too often tbe manage! must beat the responsibility for the data dump or the rambling dralt that lands oo rro deik The sadhai meat new analysts realty do not know what they are writing about They areitle given loproblem compounded by the fact that most title* are coo-stntcted more with an eye to snagging theinterest than conveying the substance of them convinced that more papers go wrong for thi) retain than any other.
The solution iaeach the analyst that the first step is to identify sprerocaUy what tt ii he rf writing about. Introducing the conceptifference between an intelligence top* and an intelligence issue is helpful in this regard An intelligence topicroad ovestion of interest, such as Soviet activity in the Third World. An imeiligence trsueevebprnenf or something ne* and different that narrows tbe torto andocal point to lhe paoer Thereimple test, an issue phrase willense of change or movement oropee not
Examples may help clarify this subtle, but Important. dKunctWr. Stno-Sovieto rntelllgence lootc butot an issue- The supuncanoe of China* espanaog economic relation* with the Soviet Union for Western in.ee torl u. Chma. or the implications of the Coibachev snccessaoo lor Soviet pobey toward China are iss-ies. Sino-North Korean relation* it an intelligence topic, the improvcmeal in Soviet-North Korean relations and what it means for China It ao intelligence issue
Tbe purpose of making this fine distinction t* to get the analyst to stop andabout what beattempting lo do before be attempts to do it The new hire Is not going to be able to make ibis dfciiwtMn; the ability to identify totelbjmw* issues is one ot tbe thing* that separates thranalyst from the adutt of the species, llblhe managerknows exactly what he is working on.
Step iwoi identify the questions. An intclnficr.ee issue is still too broad to prrnirlethe new jnalvst much help. He need* mr^lmUO guide himuhe reads
One answer is to break the intelligence tttoeeries of general Oues-tiotu. To dohe new analyst should be encouraged to step into tbe pcUcvmaWs shoes and atk hiro*eH- whatant/need to know about this issue? The questions should flow from the iMetllgcnce Issuci if they do not. the purpose of the paper is probably not deal. Using th* Sino-North Korean issue
Martin Petersen
u anolicymaker probably would wantnow: bave warmer Soviet-North Korean relation! cantedorth Korean relations orore complicated than lhal; are the Chinese concerned, what ttept hat Beiang taken to change the situation, it the Chinese leadership divided on ihe mi; -hot would China like to have happen, whatooing doing aboutorking, what do tbe Choose espert from the UntiedTbe mst cut at this shouldporittneoul. stream of cnmniootness eaescste The analyst can then weed and consolidate the fast
The list of Question* serves to sharpen tbe focus of the paper. The new analyst cow knows "whatn" and "what'sor Instance,eeds lobe coracoised about Soviet-North KorctD ccent-nrc relation) Tbe list also tefls hrro what he should be looking foi as he reads bins, and once thai at dune, il helps him identify tnleiligeoe* gaps and write reqinicmcnts Because tb* anatvsl now knows what information is relevant, il dtould also speed up the research and prevent the mdiscrirninate collection ofhe quorions may change as the analyst does the actual research, but this onlyo define the paper more precisely.
As with Identifying tbe intelligence issue, the manager willajor input in identifying the questions, and the new analyst should be encouraged to touch base with his counterparts at State. Defease, and efcewhere But. it is ImrsoriaM that tb* analyst take the first cut and actually put the Questioiis down on paper The caerciw furlW. three goats- rl sets Ihe analvsl thinking in terms of an audience, rf heightens tensitiiity to pobev relevance, and It gets Ihe analyst winking In terms of somethm, besides what happened
With the questions at hand, the analyst can do Ihe research
Step three: after completing the research, identify the two or three key points Ihe poUcvmakei is to lake away. This is the most Importanthe process. Al this point, the task for the newo:rgesl hisecide what he knowj,ut downhort paragraph or as bullets the two or three key ideas to imprest on Ihe reader Point three it the analytical bottomhe essence of tbe paper and probably the heart of the isrospeets or outlook section. If an analvsl cannot sumrnarrtc concisely but bottom line, be has not don* the analysts If be starts to write before dsHeinttnirut hit bottomhope thai tbe conclusions will fall oul of thealmost certainly will never have one
It It. ofery big step from what youo what itow do we get tbe analysl Ihere' The analvsl cannot get thee* unless he first decides what It is he knows The manager's function, then, is lo get the novice to answeronly in hisquestions outlined in step two
ava levmlumn.mw la *Kl>w4
"pJntHnlrair r
Thai done, the analyst isosition logo beyond Ihei about what the answers mean Th* preferredo use questions to think the issue through, question! designed to bring out the implication, of
Do* It Work?
To andhe four-step process will not male bad analysts adequate But it rloes help the learning process
Itommon framework and tans-nag* for managers and new hares.
Bv helping new people thinl about the process of writing fmtshedit improves iheir ability ta mastero art rathetcience.
It can be used to explain to analystsarticular drad Is deficient and offers guidance on how toick draft.
It aeu the rrew analyst focused on the consumer and OS poltcv questions
It stresses that intelligence rs interpretation of fact, not the recitation of tact..
The merits ol Ihe system outlined herothe manager, and especially tbe new nuoager. needs to develop his/her own pbn (oi traininghe art of analysis is, or should be. second nature by the time Individual, are tapped for managerial positions It may be particularly difficult to communicate to someone fust bow you do it, unless you tale the time to refect on what works for yoo and how best to get those ideasystem makes tbe manager job easier. II tUrdarrnies training aero the untt. and it allow, the manager to test different approachesystematic way
Whatevei he develops, the manager mastumber of other things to make the tool useful It HejpeeiaBy rnwortaW lo discuss the difference* between academic writing and mtelltgence with the ne. analyst, and by out for the new htre what the manager looks feeood ptece- This personal philosophy of mt-SigenceoWion that tacie itethod in your madness aad that not all you do is managerial capnetmunes The discussion alto serves to establish the standards vou will hold the new analyst le. Memo* are fine, but Ihe new Mre needs to know what the manager thinks.
Tlteie is no substitute for practice Theew hire writes, the sooner he will master Intelligence writing This ha* to be coupled withcareful reading of the finished product, but for stvfc and orgaratatton rather than substance Tbe new analyst should be given esampln, of particularly good papers, and the manager should discuss with the novice what males tbe pope, ewrsliooal
*Cwm !o4 htooili(t
* Tkt witto, mm -anay iv-uxalr.rbtxm -xu snwabir
Correctly bandbng tbe first paper is also critical. The manager should go over eacbof hi, editorial changes with the new hire, examine clearly why each was made. This may be more guidance than tbe analyst wants al limes, but It is an essential part of the teaching process.
Managing/Teachingalyits
TeocMng
Finally, 'be muiuei has to CirCfr up (rum hi* Own pvsche (hr loiw sup pressed feelings of frustration and anter that accompanied hit own learning process. Uniformly, out new hi ret were very successful in college, doingthat looks very much lika what wedo In many cases, you will be the finl person ever to tell them (heir work does not measure up. Itruly traumatic erjstrlence for many when that first draft comes bock Some will not even hear you at firstgood reason Ior also putting your comment* onyou start explaining the changes they arc too busy wondering il theyailure or ate about lo be fired. In this situation, there is no substitute foieinforcement, patience,ile wdictous handhoiding. Sues* what i* right anday lo say that theirypical and not all that different from yours. Il is at this point that tbe framework for analysis can be helpful in pointing out what works, what does not. and why.
A little luck helps, too
Thu ankle ts dunfitd
Original document.
Comment about this article, ask questions, or add new information about this topic: