W iJVVW iJl1 J1 mwvv iJ I W lrlDWl1 lSlDW lS LstiJl lS f tiJW I1 ml lBm IlU JI1 J JJ APR-JUN 1981 P L 86-36 86-36 t s U JU QE cui THIRD PARTY RELATIONSHIPS U o L - ' S r THE POETS' CORNER U o oooooooo ooooooooooooooooo oooooooooooooooooooooo 'ooooooooo 11 NSA-CROSTIC NO 33 U ooooooooooooooooooooooooo O H W oooooooooooooooooooooooooo 12 CUSSIe CABLES U TRAFFIC ANALYSIS Specialty' 14 I Without Portfolio U 15 DIA Videocasette Program U o o 19 'fIllS BOetJM N'f eON'fAINS eOBfiWORB MA'f RIAb GLAS6 rIEI B MSA G66M 123 2 REV EW eN 1 SiP 2911 Declassified and Approved for Release by NSA on '10-'1 2- 20'1 2 pursuant to E O '135 26 vl DR Case # 54778 DOCID -4019689 'rt I' t I 3 111 'r Published by PI Techniques and Standards for the Personnel of Operations VOL VIII No 4 - 6 APRIL - JUNE 1981 PUBLISHER BOARD OF EDITORS Editor-in-Chiefo oo oo o David H Williams 11035 Collection o oo o o o o 1 Cryptanalysis o oo o o o 1 1 85555 1 49925 Cryptolinguistics oo oo oo o 1 1 5981s Information Science ooooo oo L 3034S I Language P L 1 1615 Machine support o o o 1 tS084S - - 1 tC8518s Mathematics Special Research Vera R Filby 71195 Traffic Analysis o Do Taurone 35735 For individual subscriptions send name and organizational designator to CRYPTOLOG PI 86- 3 6 oocro 40r9689 EURONFIBBN IM MI I B-AGB IPBBAD U P L e eoPle workins at NSA during the last ten years have been able to witness and perhaps experience a phenomenon u likely to occur only once in the lifetime of an organization During this period NSA has moved along th t portion of its life cycle from a new continually growing organization toa middle-aged organization with a static or even decreasing work force This change was so subtle and occurred over such a long period of time that many may not have even noticed that it was taking place To those who prepare and read the internal job announcements it should have been obvious To the managers identifying and recognizing its occurrence should have been eraeial When an organization moves into the mid- dle-aged phase of its life cycle aanr characteristics surface that are quite comparable to those of a person entering middle age As with some people a middle-age spread develops In most organizations this usually occur s at the lower supervisory higher worker level G ades 11 or 12 at NSA All of a sudden the majority of the working people have achieved that grade A bulge or overabundance of people at that pay level occurs--in the neighborhood of 5000 in the Agency at present The input grade or grades below the overstaffed level suffer a corresponding decrease since there is reduced hiring and promotions continue Promotions out of the overstaffed level in this case to Grade 13 are not sufficient to reduce the bUlge since promotions in are more numerous that promotions out U Thus we have a situation somewhat similar to that which occurs when a person cheats on a diet a wide spot is created at the middle This diminishes the desireability of anything less than a supervisory position as personnel jockey for positions with the best promotion potential People who actually do the work become harder and harder to find since the rewards to be gained are not as easy to come by for the non-supervisors Specialty jobs become difficult to fill Just look at the number of job advertisements for traffic analysts or engineers if you need convincing Career development is forgotten as managers attempt to hold onto what few workers they have especially the good ones Closely guarding these valuable resources becomes a major managerial function sometimes without proper consideration of the organization as a whole or of the individual This in turn creates a frustration for those who can't achieve a supervisory status and are relegated to a worker position This frusration causes them to give up become nonproductive and to hang on only long enough to find a vacancy they consider more advantageous to their career This process creates trauma within the organization U The individual entering middle age tends to suffer the depressions frustrations and fears that are a normal part of the aging process the middle-aged organization is susceptible to these same emotions Managers are particularly vulnerable Some must make adjustments and some may have to make complete revisions in their style of management In the early formative years there is a steady stream of new bright eager employees continually entering on duty Mistakes in planning and motivation and in the managing of these employees were easy to overcome because of the continuing new input on the one hand and the opportunities caused by constant upward movement of personnel on the other Managers were certain that they could always replace a lost or unmotivated employee consequently less management of personnel assets was required and less was done Thus the manager in the ew or growing organization Apr-Jun 81 CRYPTQLOG Page 1 fIBRN't'IAL 86-36 UNCLASSIFIED needs fewer managerial skill in order to get by As the organization grows older and finally reaches the point where few new employees are entering on board a whole new type of managerial perspective becomes necessary The replacement of personnel who move up on or out becomes increasingly difficult If the manager is assigned to an area which is in the minds of the on-board employees a less than desireable one his ability to recruit from within the organization is diminished Mistakes that were previously covered up or glossed over now can result in the loss of a valuable resource The manager must now face up to his problems and make decisions More importantly these decisions must be ri ht the greater part of the time The manager must motivate his people and create a reputation for fair play and excellence which will help him in the recruiting process since recruiting situations will be such that frequently the manager must sell himself in order to attract the numbers and types of people required The manager must now learn to manage U He must motivate provide job enrichment and develop his employees so that job satisfaction and consequently morale are high within the organization Since he will usually be short of resources he must learn how to deploy them so as to obtain the maximum output from the available resources The welfare of his personnel must occupy a major part of his workday No longer cari the problems of the employee be given-short shrift ' since the loss of even a single employee can cause a severe jolt to the work force Continuous failure will eventually lead to the demise of the organization since empty billets are the most logical candidates for reduction After all if you've managed to get along without some functio for a period of-time it probably wasn't worth doing in the first place U The managers of a middle-aged organization have to lead rather than operate from a position of arbitrary authority since other opportunities are readily available to those at the working level Those who master the techniques of leadership will have little trouble attracting personnel those who can't will always be short-handed or will attempt to institute a policy of limiting intra-organizational transfers The recent moratorium on transfers of clerical personnel is an example of this When this happens the organization loses the inevitable results are discontent decreased productivity and eventually the loss of employees through attrition U In short managers of middle-aged organizations must firstly recognize the symptoms and then take whatever actions are necessary to alter their managerial philosophies so as to become leaders Some will not have to change at all others will make the needed changes and the rest will fail SOLUTION TO NSA-CROSTIC No 32 THE JOYS AND FRUSTRATIONS OF SOLUTION TO NSA-CROSTIC No 32 The Joys and Frustrations _of P1urai Dropping Arthur J Salemme CRYPTOLOG January 1978 When our daughter and a friend of hers were both five and were eating lunch at our house my wife gave them'each a sandwich and a slice of American cheese cut in strips The little friend particularly enjoyed the latter and asked what they were 'Cheese ' she was told 'Well then ' she asked may I please have another chee ' This field of research is so virginal that no human eye has ever set foot on it --Opening sentence of a doctoral dissertation in physics cited in smithsonian March 1981 Apr-'Jun 81 CRYPTOLOG Page 2 UNCLASSIFIED - -- -------- ----_ _-- ---DOCID 4019689 SECRET SPOKE II GRADING THE RUSSIAN PQE 1053 S ll 1 ince the Russian pro fessional Qualification Examination PQE was instituted in 1967 as a means to identify the Agency's most competent Russian linguists the number of people taking Part IIA the classified translation section given every six months has increased from 16 to almost 200 This drastic increase in the work load for the members of the PQE committee chosen to prepare and grade each test is one of the main reasons for the recent change in the grading system introduced last September in PQE #21 This article will explain the general testing procedure and the grading system currently used by the committee The examination itself is generally about 70 to 80 lines long and consists of four to six selections of voice and printer traffic some in Cyrillic and some in transliteration The committee attempts to ensure that the items used for the test are representative of the different types of material encountered b Russian linuists at NSA l In selecting these items the committee looks for linguistic or technical elements that a capable Agency linguist with several years of experience should reasonably be expected to be able to resolve garbles in printer traffic transliterated Russian transcription errors poor grammar or incomplete sentences in the original text omission of punctuation and prepositions highly colloquial expressions and so forth An aspirant is expected to produce a translation that accurately conveys the content of the original in good English while adhering to NSA translation guidelines Four hours are allowed for the examination E6 668 The PQE IIA sub-committee consists of a chairman and four other members all of whom are required to be certified Russian linguists from various operational o elements of the Agency The chairman ensures that the committee like the test is representative of the diverse analytical elements exploiting Russian traffic The committee selects a number of items it feels are appropriate for the test and drafts model translations Then several certified linguists who are not on the committee are asked to take the test in order to establish that its length and level of difficulty are acceptable This validation group also aids in establishing a standard by which the performance of the aspirants will be judged Next the test is reviewed by the Language Career Panel's Standing Committee on Language Testing SCLT to ensure that the test is fair as well as comparable in length and difficulty with tests in other languages It also approves the grading system to be used by the committee Finally several certified linguists who have volunteered to act as controls take the test along with the aspirants and under the same conditions U For the four tests immediat nr e e ing PQE #21 each me ber of the qo t -36 was responsible for grading several of the test items but not the entire test Three committee members graded each test item in order to obtain a more balanced evaluation These three individuals would eactI-deduct points according to a fixed numberical scale for the various types of errors l with a total score assigned to the item being graded These three scores were later combined to produce an average grade for the item and the sum of these grades was used as the aspirant's overall grade for the PQE Finally the test papers with the least number of points deducted were passed with the exact number of passes being determined by the control group The passing mark for the aspirants has generally been set at a significantly lower level than that of the validation and control groups U This approach while satisfying the need for a reasonably objective grading system led to several problems which were noted by both the aspirants and the committee members To begin with the large number of test papers involved forced the com- Apr-Jun 81 CRyl tOLoG Page 3 SBCRET SPOKE P L 86-36 DocrD 4019689 SHeRH SPOKE mittee to spend an enormous amount of time on establishing an exact grade for each paper--six months for PQE '20 as an example This seemed to be a less than efficient system since most of the committee's time was spent grading the papers with the most errors and the least chance of passing It should be noted in this connection that some PQEs in other languages have been graded using the same system of deducting points for errors but in conjunction with a cutoff score such that anyon losing more than an established number of points automatically failed the test making it unnecessary to continue grading the paper Furthermore the number of aspirants for the other languages has been much smaller as few as two or three in some cases which meant that less time was required for grading ti The numerical grading system as used for the Russian PQE has also inadvertently tended to maximize the attention given to minor errors since a grader under pressure to arrive at an accurate score for each test is inclined to look for minor errors and to deduct points for them This has unfortunately at times resulted in disagreements and grading discrepancies among the committee memebers since no two graders are likely to share the same opinion of the boundary between poor word choice and a definite mistranslation or of the serious ess of poor spelling and grammar in a translation In addition the numerical grades while providing apparent objec ivity have tended to emphasize the number of errors and to obscure the nature of the errors one 4point syntactical error may mathematically equal four I-point English usage errors but these cannot be considered equivalent reflections of a linguist's ability to produce an acceptable translation This caused some problems during the grading of PQE '20 since the length of time required to grade the tests in the first place precluded any comprehensive review of the grades before they were released U As a result of the problems arising during the grading of PQE '20 the committe preparing PQE '21 decided to modify the grading system to incorporate the cutoff feature mentioned above The cutoff point like the curve used for grading the preceding exams was to be based on the performance of the validation group It was also decided that the committee would no longer attempt to assign an exact numerical grade to each test Under the new system a grader was to read the entire test paper and assign a Pass or a Fail to each test--as had been done for PQEs #15 and 16--after comparing the aspirant's performance to those of the members of the validation group with the emphasis on the major errors committed by the aspirant Each test was graded by three committee members and a unanimous opinion whether pass or fail was recorded as such without further discussion Split decisions were resolved by discussions among the graders with the final grade determined by consensus Minor errors were generally taken into account only for borderline papers which were failed if they contained an excessive number U The new grading system has already shown itself to be superior to previous methods in several ways First the evaluation of the tests was completed in just three months as compared to the six months required for PQE '20 Most of the committee's time was spent on borderline papers where more effort was required rather than on the papers that would clearly be failed in any case Furthermore the emphasis in grading was directed to the most serious types of errors with a corresponding de-emphasis in the more subjective areas of word choice elegance of phrasing and the like Finally the immediately apparent distinction between a Pass and a Fail --as contrasted with the difference between two numerical scores--ensured that any disputed tests would automatically be reviewed by the committee with any grading errors such as overlooking omitted sentences being detected during the review process U The rest of this article will provide examples of the principal types of errors and show the relative weights assigned to them in grading The examples are given either in Cyrillic or in transliteration just as they appeared in the original material from which they were extracted Also garbles and other defects in the original have been left uncorrected Each example includes the original Russian R the roodel translation prepared by the ccmni ttee M and the aspirant's version T f6EURt- MAJOR ERRORS The focus in grading as explained above is on major errors-those for which four or eight points would have been deducted under the previous system These can for the most part be regarded as errors in grammar or syntax The aspirant has not understood the relationships among the elements of a sen- Apr-Jun 81 CRYPTOLOG Page 4 SREURHT 8POKI1 DOCID 4019689 SECRE SPOKE I I tence or among the sentences in the orlglnal and has accordingly produced a translation with serious distortions of the meaning and structure of the original The triage system used for preliminary grading separating the papers into pass fail and needs-further-discussion categories used the number of grammatical and syntactical errors as its primary criterion although certain other errors such as the omission of an entire phrase or sentence were also considered to be major EO 1 4 c F L 86-36 I Apr-Jun 81 CRYPTOLOG Page 5 DOCID 4019689 TOP gJijCRBT UMBRA EO 1 4 c P L 86-36 expected i this type of error occurs much more frequently than grammatical or syntactical ones This may to some extent be a result of the four-hour time limit on the PQE but in many cases it appears to be more a result of carelessness or indifference on the part of the translator Other errors treated as minor by the PQE #21 committee have included the omission of a word in the translation extremely awkward or incomprehensible English incorrect verb tense usage and so forth Here are some examples of minor errors from PQEs #20 and 21 The aspirant treated an adjective as a noun and separated it from the noun it modifies It should be noted that there were several errors in the transcript used for this item masculine and feminine modifiers for a neuter noun and that the aspirant's error may have originated in his mistranslation of the verb SCHITAL In summary most of the common major errors involve verb forms Russian word o der colloquial expressions and noun endings These in addition to lexical errors can lead a translator to see nonexistent relationships in the text and to produce translations based on them sometimes causing distortions in other parts of the text so as to make them fit the translator's perception 'ffSEt- MINOR ERRORS Lexical errors are considered minor only in the sense that they are less serious than grammatical and syntactical ones Such errors for which two points were deducted under the Child system cannot however be considered as insignificant Gross mistranslations can occasionally affect the meaninl of a text as seriously as a syntactical error and even less severe ones will degrade the quality of a translation As miKht be Apr-Jun 81 CRYPTOLOG Page 6 OP t'HCRH tJMBRi DOCID 4019689 EO 1 4 c P L 86-36 TOP SECRET UI'fIBftA account In spite of objections from several aspirants that the purpose of the PQE is to test command of Russian not English an individual translating from Russian into Ellglish is expected to be able to express himself well in the latter Aspirants are encouraged to bring an English dictionary and a thesaurus to the examination along with the Russian-English dictionaries recommended by the panel U The area of word choice and phrasing is more ambiguous Graders are usually fairly tolerant at least under a Pass Fail system as long as the translation can be understood Mistakes like beginning a formal intergove rnmental message with Dear Friends inappropriate style will generally be overlooked by the committee in determining an aspirant's grade Aspirants should realize however that a translation that disconcerts a reader or forces him to make an effort in order to understand what was mean--such as might result from adherence to Russian word order--is an inferior one NSA TRANSLATION CONVENTIONS Aspirants are also expected to follow the minimal format rules prescribed for NSA translations by USSID 300 capitalizing proper names indicating interpolations by the translator noting garbled words and so forth As has been the case with style and spelling errors no aspirant has failed the Russian PQE exclusively because of failure to observe Agency conventions It is however recommended that aspirants familiarize themselves with the relevant portions of USSID 300 before taking the test and attempt to follow these conventions as much as possible 1 1 j For PQE #21 this type of error was taken into account primarily for tests near the rough cutoff level designated for the number of major errors A test that would have been borderline on the basis of the major errors it contained would fail if it also had a large number of lexical errors with the cutoff point for these minor errors being determined by the performance of the validation group U ENGLISH ERRORS Although no aspirants have been failed solely because of poor spelling or English grammar--errors for which one point would have been deducted under the Child system--these mistakes do affect the quality of a translation and may influence a grader's opinion of a paper even if not formally taken into U CONCLUSION The Russian PQE IIA committee feels that the modified Pass Fail system now being used is an improvement over the point system which it replaced In addition to requiring less time for the grading process--a vital consideration in view of the large and ever-increasing number of aspirants--we feel that it can provide results that are both fair and reliable 1 For a description of this scoring system see An Objective APyroach to SCOrij T r a n slations by CRYPTOLOG March 1976 Apr-Jun 81 CRYPTOLOG P ge TOP SECREI' UMBRA 7 P L 86-36 --OOCID 4lJl9689 SBEURRB'f Third Party RelationshipsM 1 L 18309 Apr-Jun 81 CRYPTOLOG SFlEURRB'F HAVE THEM WHAT ARE THEY DO WE USE THEM PROPERLY ARE THEY WORTH THE PRICE IHY Page 8 l------------- ----- P _i EO 1 4 c EO 1 4 d L 86-36 ------ DOCID 4019689 SEeKEr J 1 EO 1 4 c EO 1 4 d P L 86-36 Apr-Jun 81 CRYPTOLOG Page 9 SECRET IIANBbB fA S8MIN'f SlIAtHIBbS 8UbY DOCID 4019689 TOP SECRET UMBRA DEPARTMENT OF CLASSICS AND MYTHOLOGY U EO 1 4 c EO 1 4 d P L 8 Is there any mythology buff out there who can identify whatever personal attributes of these two Greek goddesses make this quotation an apt one If so please all CRYPTOLOG on 1103s 6 S 6------------------------------- ------A r-Jun TOP 81 CRYPTOLOG Page 10 SECRE UMBRl 86-36 EO 1 4 c P L DOCID 4019689 eONFIBENTIl 1 U Introduction is e eee Poem is UNCLASSIFIED In the mid-l9S0s a cryptanalytic effort was underway in ADVA the predecessor of today's AS Cipher Analysis against a particularly recalcitrant wired-wheel system The R D element then known as MATH comparable to the current R5l Mathematical Research Division as part of its mission to provide applied mathematical support to PROD elements suggested to ADVA a novel for those times approach involving eigenvectors The ADVA analysts resisted this as being no improvement over the techniques they were already using MATH reported this reluctance to the ADVA leadership slyly adding that perhaps ADVA was frightened by eigenvectors A lively discussion ensued the upshot of which was this D F sent from William Lutwiniak Deputy Chief of ADVA to Arthur Levenson the Chief Apr-JllJ1 81 CRYPTOLOG Page 11 EURONPIB6N'fIAb II NBbB '1 8SMIN'f 8II NNBb8 SNb'I' DOCID 4019689 UNCLASSIFIED NSA-Crostic No 33 L 86-36 D H W Apr-Jun 81 CRYPTOLOG Page 12 UNCLASStFIED 4 1 DOCID 4019689 UNCLASSIFIED 86 36 Apr-May 81 Page UNCLASSIFIED 13 DOCID 4019689 GLAIIIC GABLBIC U 1 eelr P r o o -r----r 1 1ooo oooooo II oo 1- I o oo oo o ooo-1 o - o oo e o - I Most people find that REQUEST STATUS OF RESPONSE TO REF MESSAGE is sufficient But now and then there are exceptions P L 8 Apr-Jun 81 CRYPTOLOG Page 14 HAtJ8 bB 'qz EURJQJcUNT SIb' NNEbiS Q JbV DOCID 4019689 StJEURRB'f This is an address given by Mr - ltheA2SeniorAnalYst at the 1980 Symposium of the Communications Analysis Association and is based on his extensive experience on the Soviet Military Problem TRAFFIC ANALYSIS Specialty Without Portfolio I IA2 ------------ -- -- eU SECRET This article is classified Ilfdoffii ViA MoffItt' ffd 4fM ONLY in its entirety n addition to cryptanalysis which I do not intend to deal with in this discussion communications intelligence has always involved two primary overlapping but distinct tasks with respect to target analysis I ordering and understanding target communications structure and procedures--commonly referred to as trafftc analysis and cdescribing target organizations activities intentions and trends --known as intelligence analysis or special research analysis c The major overlaps of the two disciplines are in their contribution to target organization or order of battle and command and control Of these two skills only the latter one intelligence analysis has been effectively accommodated in the NSA career structure Traffic analysis in terms of its application is largely regarded as a skill to employ if time and resources permit which means that it is basically nonfunctioning in many important areas In career terms it has been essentially inaccessible to representative numbers of people being an apprentice slot which must be abdicated by those with more than modest ambitions i Apr-Jun 81 CRYPTOLOG Page 15 SECRE'f EO 1 4 c P L 86- 3 6 Ut LlI E I' SQAIIN'T 811ANNBhS 6141 DOCID 4019689 EO 1 4 c P L 86-36 SRCRE'f Apr-Jun 81 CRYPTOLOG Page 16 SHeRe Ut l'IBhEl 'AA S8ldlU'f' 8ItAlmBhS el4bY DOCID 4019689 EO 1 4 c P L 86-36 SEEURRET Apr-Jon 81 CRYPTOLOG Page 17 Il1dfBbl l 'f1l 86MHff ellAfRfBbS elf 'I DOCID 4019689 SECRET ATI'ENfION PLATFORM USERS U Those readers who wish to send contributions to CRYPTOLOG and who have access to PLATFORM may forward their submissions via PLATFORM Send them to cryptolog that's all in lower case at BARlC 5 Contributions naturally will also continue to be accepted in handwritten or typed form Any questions Call 11035 s A pr-Jun8l CRYPTOLOG Page 18 EO 1 4 c P L 86-36 SECRE DOCID 40196ff9 CO J'II TIlL j 1 DEFENSE INTELLIGENCE VIDEOCASETTE PROGRAM U Did you know Collateral Documents T5322 holds a collection of videocasettes which are products of DIA's Defense Intelligence Videocasette DIV Program The subjects range from documentaries on the Soviet military to interviews with authorities on Chinese society U The following is a list of the titles currently available All titles are UNCLASSIFIED Classifications of the tapes themselves vary The Soviet People The Soviet Soldier Soviet Airborne Forces Soviet Amphibious Forces Soviet Air Operation Concept Soviet Tactical Electronic Warfare The New Soviet Tanks Soviet Armed Helicopters The KIEV Deploys BACKFIRE FOXBAT A j 1 The SS-20 The Soviet Afghanistan Invasion Warsaw Pact Military Posture The Chinese People China's Military Posture U Except for those titles marked with an asterisk these tapes may be borrowed from T5322 for a period of two weeks For further information caq Ion 8642su Themasteriskedmtitlesuumay be obtained from A2045 5091s Apr-Jun 81 CRYPTOLOG Page 19 i 1 C9NflJ TIAI P L 86-36 DOCID 4019689 UNCLASSIFIED HELPl CRYPTOLOG has no magical automated way of updating our address list So if you change jobs without letting us know your continued receipt of the magazine will depend on the distrib tion personnel at you former job Soem are very helpful in calling to notify us of new addresses But not all are Also if your job doesn't change but your organization does because of reorganization you may not continue to receive your copy So if you have moved or if you have been reorganized the best way to insure that you will continue to receive CRYPTOLOG is to give us a call at 1103s Apr-Jun 81 CRYPTOLOG Page 20 UNCLASSIFIED Pl-oct 81-83-9'5' This document is from the holdings of The National Security Archive Suite 701 Gelman Library The George Washington University 2130 H Street NW Washington D C 20037 Phone 202 994-7000 Fax 202 994-7005 nsarchiv@gwu edu