U lDVOl JU lDl1 L'UlWV $lBl JU10V lDffi $U lB UlW OO Dfi f W DW l1 DUJfi lBl IlU JI1 J WaJ 0 0 tJ PROJECT HELIPAD AN EPITAPH U oooooooooooooo o ooooo 1 TRANSCRIPTION TRIO I WHERE DO GOOD TRANSCRIBERS COME FROM U o o oo oo 4 I I RUN THIS THROUGH YOUR TRANSCRIPTION MACHINE U 6 III TRANSCRIBER - ANALYST RELATIONS U oooooooooo ooo Doris Mi11er oooo o ooo o 8 CISI NEWS 1979 SPRING CONFERENCE U oooo o ooo ooo o o ooooooooooo oo o o ooooo 9 CLASSIC CABLES U oooo ooooo oo ooooo oo oo o oooo o o o oooo ooooo o oo o 10 EMITTER IDENTIFICATION TECHNIQUES U ooooooooooo o ooo 1 ll MORE THAN WORDS CAN SAY U J Gurin o ooo o oo oo oo 13 NSA-CROSTIC NO 25 U oo oooooooo ooooooooo oooo o ooo D H W oooooooooo ooooo oo 16 MORE FAIRBANKS ON ENGLISH U oooooooooooooooo o o o Sydney Fairbanks oooo 18 CHAPENKO SHAPENKO--WHAT DIFFERENCE DOES IT MAKE U o t 19 U t o IIIS BOOBMRN'f OON'fMNS OOD WORB MA'f eRIA GLASSIFIE9 BY NSA GSSM 123 9EGLASSIFY eN 1 MJW 2BB a 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 4009-B27 Published Monthly by PI Techniques and Standards for the Personnel of Operations VOL VI No 5 MAY 1979 WILLIAM LUTWINIAK PUBLISHER BOARD OF -- EDITORS Editor-in-Chief David H Williams 39575 Collection I I 555s Cryptanalysis o o 1 1 49025 Cryptolinguistics o o 1 Information Science 1 59815 1L - Language o o o 1 HS7Us Machine Support 1 Mathematics ' 1 Vera R Filby 7ll9s Traffic Analysis Don Taurone 35735 Production Manager o Harry Goff 52365 For individual subscriptions send name and organizational designator CRYPTOLOG PI SHCRHT P L S 4Sj 1 85185 Special Research to ' o It8161S 86-36 DOCID 4009827 SECRET SPOKE P L 86-36 Project HELIPAD An Epi tap h I EO 1 4 c P L 86-36 _ U _ _ IJI Jlurin g th' Vi'tnam confiict USAF pilot' as aggressors these pilots even took on the trappings of a Soviet fighter unit flyi g Soviet flags wearing Soviet insignia and decorating their offices and ready rooms with Russian wall posters manuals photographs and model aircraft Their own fighter5 carried Soviet markings UI That I s the beginning of the story in Soviet air combat tactics at least to the extent that these tactics were known Initially these pilots flew the T38 TALON and later the FSE TIGER II to simulate the Soviet MIG-21 FISHBED and MIG-23 FLOGGER fighters Their mission was to train other USAF fighter pilots how to fight and win against Soviet aircraft and tactics Known May 79 CRYPTOLOG Page I g CRE'f SPOKE EO 1 4 c P L 86-36 DOCID 4009827 SECKEl' SPOKE May 79 CRYPTOLOG Page 2 8 8CR 8'f SPOKE EO 1 4 c P L 86-36 DOCIO 40 09627 EO 1 4 c SECRE'f SPOKE P L 86-36 I 4 I PROFICIENCY CERTIFICATION FOR LINGUISTS U U The Language Proficiency Test certificate program for military linguists is now under way The first certificate has been sent to lufor passing the Spanish LPT I I U Other certificates are being printed and will be presented in the near future If 'oii are interested in getting more information on how to receive a certificafe for your linguistic abilities contact CAPT USAF N36 phone 5578s or 688-6565 May 79 CRYPTOLOG Page 3 SKCH'f SPOKE P L 86-36 DOCID 4009827 SECREJ' Transcription Trio I Where Do Good Transcribers Come From In the fr ZOl rina ar rle the first of our' authors answel'stheabove question y saY1-ng hey don't Good transaribers are the result of a lengthy period of alasSl'oom and on-the-job training right here on the premises he Russian voice transcription effort II at NSA has been undergoing a gradual but ar-reaching evolution over the last five As stated previously the Agency's transcribersare not the product of a common language background Some have received strictly military language training some have college degrees in Russian and still others with either high school diplomas or college degrees have received their Russian training here at the Agency Each of these training methods has its own unique strengths and weaknesses in order to compensate for some of these weaknesses the NCS provides aplethoTaofcourses u The courses offered by the NCS are designed to address the transcriber's specific language problems Courses are offered in the fields of grammar syntax linguistics phonetics translation conversation and area studies These courses are intended to fill in the gaps in an individual's language background and to allow him the opportunity to improve his skills on a continuing basis By taking these courses the transcriber i able to update and refresh his knowledge the Agency by sponsoring these courses during the work day indicates its support for this on-going process Many transcriber a lso improve their skills by taking dvantage of the Agency's After-Hours S udy Program This has proven especially beneficial to the m ntary-trained transcribers who are able to obtain college degrees through this program and increase their understanding of the Russian language and of the Soviet Union as a whole u The Agency also provides numerous opportunities for those individuals interested in making transcription a career The first step for most is the Agency's professionalization program To be certified as a linguist an individual must demonstrate a high level of competence in both translation and transcription Serious career linguists also have the opportunity to attend the two-year intensive language course taught by the Department of the Army in Garmisch West Germany A recent exciting development for some transcribers has been the decision to allow a small number of career transcribers to progress beyond the Grade 12 level without the normally attendant managerial responsibilities This step was taken in response to the critical need to retain senior transcribers in transcription billets these individuals will thereby be able to devote their full time and talent to that which they can do best--transcription May 79 CRYPTOLOG Page 4 SBCRBCf IIM1BloEJ ViA eSMHf'f eIlAt cBhS 6141 1 P L EO 1 4 c P L 86-36 86-36 I I - DOCID 4009827 EO 1 4 c P L 86-36 SECRET you'll be lost trying do do it in Russian Too often in the past the Agency has invested money and years of training to produce a highly competent linguist only to put that person in a managerial job where his talents cannot be utilized fully esufficient understanding of the Soviet society to comprehend the personality and habits of the person speaking I i U It becomes obvious therefore that if a transcriber is to produce a finished transcript he must gain at least a ifundamental knowledge of the subject under discussion There are two main areas of knowledge which every transcriber must learn regardless of his proficiency in the Russian language itself The first such area could be termed target knowLedge Target knowledge is comprised of three quite different parts evocabulary egeneral understanding of the subject involved if you don't understand the concepts in English May 79 CRYPTOLOG Page 5 SBEURRBtF EO 1 4 c P L 86-36 WUIQbK HI EURlQJoUNl' EURlHMltIEbS SUeY -DOCI-D 4009827--SECRHI' Eo 1 4 c F L '86-36 in the analyst's understanding of what is actually taking place U Still another essential factor in the training of any transcriber is his ability to use the multitudinous transcriber working aids A transcriber's working aids might be compared to a physician's medical reference books regardless of the individual's innate intelligence his ability to handle a given set of circumstances would be severely hampered if not made impossible by their absence U The extent of the resources available to a transcriber is such as to bewilder the trainee Over a period of years many transcribers gradually accumulate a considerable number of their own personal reference materials These working aids may take on any number of forms from printed dictionaries to computer files addition to these basics there are other working aids which enable the transcriber to do his work most of these are specifically related to the subject matter of the material being transcribed in his area These may include the following In us first look at those basic reference books which are widely distributed to incoming transcribers Let -A standard Russian-English dictionary obviously the first item u It cannot be stressed enough that these resource materials are essential to the production of a finished transcript all of them many of which originate within NSA are worthless in the hands of an amateur but invaluable to the experienced and knowledgeable transcriber II Run This Through Your TranscriDtion Machine u I I P L I t is sometimes assumed by non-linguists that transcribers are sort of like parrots They simply repeat in written form what someone has said Thus while a transcriber needs to know the language of what he scribes he certainly doesn't need to think just repeat This is of course a very narrow view of the transcriber's task at NSA and while you might laugh and say I know there's more to it than that it is quite surprising to find the number of people who feel that transcribers are like machines An inside joke among transcribers is the remark attributed to a nontranscriber who once asked one of the transcription offices to run this tape through your transcription machine -I May 79 CRYPTOLOG Page 6 SHCRHT IIAt18 b8 VIA 8SMIN'f' 81h N141 3hS 6NhY 86-36 - 7 ' u 1 4 e F L 86-36 u If only it was that simnle o May 79 CRYPTOLOG Page 7 SECRET UANPI i HI Q II JT QII4 J I iib8 Q IbY DOCID 4009827 SHEURRHT L 6 P 86-36 CUI rhus as can be seen a transcriber is by no means a simple parrot or machine The job he is asked to do is not only quite complex but requires skill and dedication to enable him to stick with a job even when the going gets rough CUI I hope that this article has provided an insight into wh t it is that Agency scribes do and dispels the myth that voice processing is a simple matter of running this through the transcription machine III Transcriber-Analyst Relations EO 1 4 c F L 86-36 CUI Doris Miller Retired May 79 CRYPTOLOG Page 8 H NPI i HI' EURJQAIIN 8IIANNEhS Of4L DOCID 4009827 EO 1 4 c P L 86-36 CONPIBEN IAL o THE COMPUTER AND INFORMATION SCIENCES INSTITUTE U THE USER -- IT'S ABOUT TI ME or J U U Mark your calendar today for May 21-25 because you'll not want to miss the 1979 CIS I Spring Conference The Computer and Information Sciences Institute is sponsoring an exciting and informative Computer Conference with this year's emphasis on YOU the computer user The speakers who represent some of NSA's most knowledgeable professionals in a variety of career fields will be talking on topics and issues that are extremely important and topical to NSA's computer user community U With intriguing titles such as I Want My Program to Talk to Me and Wanted A French Chef for Analyt ic Programming and state-of-the-art subjects such as RASTER graphics at NSA Human Machine Interfaces User Computer Security and Intelligent Terminal Applications this year's sessions promise to be the best ever U Schedules for each session will be posted the week before the conference CISI is anticipating a large turnout for the sessions and especially wants you to be a part of this SUPER EVENT U The general topics for the five sessions are Monday 21 May The User Himself Tuesday 22 May Applications Wednesday 23 May Computer Security Thursday 24 May Acquisition Methodology and Standards Friday 25 May Tools U All sessions will be held in the Friedman Auditorium from 0900 to 1130 May 79 CRYPTOLOG Page 9 CONPIBRN IAL W't JQlsg '7 4 eQMIN SII dH1EhS SNbll DOCID 4009827 CONFIBBNTIA elbAIIIC GABLBI r o o - ooo I 1' oooooo ooo1 o o oo 1 r I o o o o oo ooo oo o e e oo 1oo o o o o 1 oooooooo DEPAR-weIT OF FULLER CLARIFICATION P L May 79 CRYPTOLOG Page 10 CONFIBBN'f'IAL IIA Nl3bE HI QQJcIl JT 1511 J IKIsi O II Y 86-36 SHCRET DOCID 4009827 Emitter Identification Techniques I IW2 P L This article is classified in its entiret he term emitter identification denotes Evaluation of such a technique is interpreted the association of a signal with a spehere to mean cific source or platform The identification evaluation of the choice of features process usually involves matching parameters evaluation of the performance of a of the signal with those from known emitters decision rule on vectors of features and producing a list of candidates Related information such as that obtained from Theissuesaddressedare'tllElf'QUQwing eusefulness of the features How wel O L 1 4 c do the features separate the known 86- 3 6 emitters _adequacy of the library for training How many transmissions of eachernJ O- 1 4 c ter are necessary for the givense't of features P L 86-36 ecomplexity of the decision rule Is the decision rule suf iciently sophisticated to separate the emitters in the chosen feature space eestimation of the error rate of the decisionpdle What is the expected percentage of misclassifications in new data _incorporation of no-decision options into the decision rule an open library versus a closed one T For over a year RSI has been investigating mathematical statistical techninues for addressinQ asoects 2 and 3 I I The RSI effort has focused on ways to evaluate an emitter identification technique There are many types of decision rule that may be chosen for use in emitter identification Among those studied by RSI are the May 79 CRYPTOLOG Page 11 SHCRH'f 8 IQr IillJfI QMIN'F 8Ib't iHl'lhS 8HhY P L 86-36 86-36 P L 86-36 EO 1 4 c DOCID 4009827 SECRE'f following _minimum distance or maxi l correlation - k-nearest neighbor weighted or un weighted _ discriminant functions linear or quadratic - Bayes' rules The parameters of a decision rule are estimated using a learning set consisting f transmissions from known emitters The In summary R51 research I finite sample performance of the decision rule on a prescribed set of features can be evaluated on a test set a different __ Ihas made available a plethora of set of transmissions from the same known techniques and computer programs that can emitters in terms of be used by those involved in emitter identi_error rate percentage of misclassification work All software runs are on the R5 CDC 6600 but most of it is in Fortran and fications should be readily available to other computers _confusion matrices correct class Questions about these technique nil or 1 h membership versus that produced by software should be addressed tol I the decision rule Chief R51 8518s orl IChief _expected position of the correct R513 8525s identification in a list of candidates _ robability level associated with the 1 Details about these algorithms and correct identification versus that methods can be found in the following of an incorrect one R5l publications Cluster Analysis Introduction to Models From test set performance one can infer and Methods NSA Technical Journal Vol decision rule performance on unknown data XXII No 2 Spring 1977 and decide whether or not the given features and decision rule are adequate for the pur Introduction to Statistical Pattern pose intended l Recogni tion NSA Technical Journal Vol XXII No 3 Summer 1977 Some R51 work has been done on library maintenance The inclusion of erroneously Some Thoughts on Coalitions Discrimiclassified emitters into the library can nant Analysis R5l MATH 28 77 20 July 1977 distort the set of reference patterns Peri- I FOR THE PUZZLE FANS U U Given below are the numbers of the squares which contain the final letters of each word of the NSA Crostic on page 16 Simply thicken the lines at the right sides of these squares and you're in business 4 8 18 22 31 35 39 44 48 50 53 57 59 63 67 72 74 83 88 90 92 96 101 105 107 112 114 120 122 128 134 136 145 149 151 155 158 164 175 177 184 188 195 198 206 209 211 217 221 223 227 233 240 244 248 255 257 260 271 273 277 281 SOLUTION TO NSA-CROSTIC NO 24 U CRYPTOLOG April 1979 by D H W I t'fThelBucky Balance CRYPTOLOG July 1978 U The best way to identify a Project Starter or a Project Finisher is to analyze the person's working habits closely The Project Starter is interrupt-driven He operates on a last-in-first-out basis The Project Finisher functions as a firstin-first-out queue Every task that enters his queue comes in at the lowest priority May 79 CRYPTOLOG Page 12 SECRET IIMiBhl3 'rlA e6MIH't' ellAHHl3hS 6Hh t P L 86-36 ____ L_ L _ UNCLASSIFIED DOCID 4009827 J GURIN R51 JItow Pfltan O f an Jay s a part-time jazz musician I can vouch the fact that there seems to be no way A offor writing down the improvisations of some 6f our jazz greats even after listening to tneir recordings again and again One can capture on paper the notes they play without too much difficulty but not how they play those Rotes The same problem plagues the classical musician as he attempts to play compositions of the masters How loud is forte How fast should presto be How much emotion in con amore and how much and how quickly should one slow down for ritard The instructions left by composers serve only as very imprecise guides Although many claims have been made for this system or that one none has succeeded in recording the dance so that the choreography of a Fokine or an Ailey may be transmitted other than by example There just doesn't seem to be any way to notate choreography successfully and even the motion picture which once was looked upon as the answer to the problem has shown fatal weaknesses as a complete record Writing systems for spoken language also present only approximate representations of what is said I'll never forget how impressed I was when for the first time I heard Mark Antony's lines Friends Romans countrymen lend me your ears spoken by a f ne Shakespearean actor It sounded very dIfferent from the dull recitation in my high school English classes The words were there in the book we read in school but something had to be supplied by the actor It is possible or course to indicate how a sentence should be read by using italics for stressed words or by underlining those words Thus we may distinguish Don't ever do thilt again from Don't ever do that again But this device only tells us that the underlined word should be emphasized it doesn't tell us how or how much Some sentences remain ambiguous until spoken aloud as in the case of Jane left directions for George to follow which can have two distinct meanings There are a number of terms that are used to describe those features that distinguish completely meaningful speech from the simple ecitat on of strings of words Prosody and ntonat n are two of the common ones while the fancy word suprasegmentals simply describes features which extend beyond segments of words su h as phonemes The easiest way to describe any of these is in terms of what it is not It is what is left when you take away traditional grammar vocabulary and rules for pronunciation of segments of speech In other words you are left with pitch loudness and duration The nature of our concern here directs us to the perception rather than the generation of these features For us there is no sound in the forest when a tree falls--if there is no one there to hear it And by introducing this notion of human perception we bring in the factor which complicates everything If it weren't for the perception mechanism and its special qualities it would be a simple matter to equate pitch loudness and duration with frequency intensity and time which are easy to measure and manipulate In fact it may seem obvious now that there is no one-toone relation between frequency pitch intensity loudness and length duration but this is a relatively recent discovery It was Galileo who first established pitch as a function of frequency and it was not until the 1930s that it was shown to be a function of intensity as well and possibly of other factors In some ways perhaps duration is the easiest to understand Its principal ingredient is time But in examining the question of the shortest durations of sound that a human listener is capable of noticing investigators have determined that intensity plays apart Increasing the intensity up to a certain level tends to improve the listener's capacity to discriminate between duration differences Loudness presents another concept that on the face of it seems simple enough but can be shown to be more complex when one realizes that perception of loudness involves some tricks played by the mind A simple ex- May 79 CRYPTOLOG Page 13 UNCLASSIFIED DOClD 4009827 UNCLASSIFIED periment displays this dramatically Assemble several radio receivers in one room tune them all to the same station then mark their volume controls at the settings that make them all sound equally loud when pLayed separateLy Listen to one radio by itself then turn another one on to the mark on the volume control The two together will be louder than one alone of course Now do the same with a third the loudness will increase but not as much as before Adding each of the remaining radios will continue to add to the loudness but each time the increment will be smaller Consider another dramatic illustration of the relationship of the intensity power of sound and the loudness of that sound as perceived by the ear and brain Under the proper conditions the intensity of the sound of a full orchestra in a fortissimo passage may be about 20 million times as great as that of a solo violin in a soft passage But of course it does not sound 20 million times as loud When one considers the relationship between intensity and loudness it is clear that one can scarcely be simply proportional to the other It seems that there is a law which is more or less true for all sensations pressure sight hearing etc --that the intensity of the sensation is proportional not to the stimulus but to the logarithm of the stimulus This means that every time the stimulus is mUltiplied by the same factor the sensation goes up one step In other words me to quote a definition from the MerriamWebster dictionary which is not quite correct According to that usually reliable source pitch is that property of a musical tone which is determined by the frequency of the sound waves which strike the ear The larger the number of vibrations per second the higher the pitch Close but definitely no cigar Frequency may well be the most important element in determining pitch but it is by no means the only one For example judging the pitch of a single-frequency sound depends to some extent on the loudness level of the sound A large increase in loudness may drop the apparent frequency by a readily perceptible amount Because of this apparent lowering of pitch a vigorous trumpet player will often accuse his gentler colleagues of playing sharp because his own notes are so loud in his ears The orchestra's conductor has to mediate this quarrel and must try to arrange things to satisfy the audience which hears all the instruments at a lower level of loudness where the effect is negligible In a series of experiments it was found that with increases in intensity there were pitch decreases for all frequencies up to 2000 vibrations per second after which there were very slight pitch incpeases as intensity increased Some of these strange goings-on may be explained in part by the fact that the ear is not equally sensitive to all frequencies Of two sounds of equal intensity but of different frequency one may be subjectively Sensation Stimulus louder than the other if it falls into a frequency range at which the ear is more increases one step sensitive goes from 10 to 100 increases one step goes from 100 to 1000 goes from 1000 to 10000 increases one step Another illustration of the relationship etc between pitch and frequency is this if a musical tone is composed of a series of harBefore you relax too much on the subject monic frequencies with which frequency is of loudness let me add the following Our the pitch to be associated Surprisingly perception of loudness is also influenced by enough it may not be the one with the most frequency For example for two tones not energy nor the lowest frequency As a matter far removed in frequency doubling the power of fact if a given musical sound were comby adding these tones results in approxiposed of frequencies at 200 300 400 and SOD mately double the loudness But doubling Hz then one would hear the pitch at 100 Hz the power by adding the same tone to itself Or if the component frequencies were at 400 may not double its loudness at all In fact 600 800 and 1000 H one would judge the pitch we may have to increase the intensity of a to be that of a 200 Hz tone If you wish to tone as much as eightfold in order to double continue the game add to the last batch tones its loudness This holds true for all except at 500 700 and 900 Hz and the perceived pitch very faint tones where doubling the power will drop an octave to -100 Hz In other words more than doubles loudness when a tone consists of several frequencies differing by a constant amount the perceived Of the prosodic features of speech the one which seems to attract the most attention pitch may be that of a tone whose frequency is pitch Here again we run into problems of is equal to the common difference as shown below So in these cases one hears pitches perception And just to demonstrate that with frequencies that aren't there sometimes old friends may fail you permit May 79 CRYPTOLOG Page 14 UNCLASSIFIED _____L __ J DOCID 4009827 I i o -- - -- _ -- - ---- _ ---- _ UNCLASSIFIED deliberately left hanging in mid-air and clearly signifying that there was more to A B C the thought than had been uttered has a rising intonation 1000 xxx xxx Granted that in speech the pitch pat900 xxx tern conveys information about the speaker 800 xxx xxx his emotional and physical state his cul700 XXX tural and regional background etc my prinxxx 600 xxx cipal concern here is with pitch that dis500 XXX XXX tinguishes the meanings of utterances The other day I heard a joke that I did not un400 XXX XXX XXX derstand because the teller failed to use the 300 XXX right pitch pattern for the pun which was 200 the point of the story The joke was about 100 a clone who had become very foul-mouthed and Actual frequency XXX was pushed out of a window on the 17th story Apparent frequency I of a building The punch line was He made an obscene clone fall As the story was told the pitch patIn tonal languages like Chinese pitch is tern for the punch line was like the one in a vital element in the lexical sense That is to say individual words contain a pitch pat- I He made an absurd man call while it should tern which distinguishes them from other words have been like the one in He made an obscene phone call Speakers of non-tonal languages find t d if- ficult to hear these tones but not because Let's face it the human mind simply they have no exper ence in ndling pitch isn't up to perceiving speech sounds exactly The Sngltsh intOftatlon systea is fully as as they are produced The ear itself is co licated as any tonal system used in Southphysiologically incapable of responding to east Asia or Africa Fwrth rmore English changes which are too rapid or too minute speakers easily learn to sing songs employThe subjective record of how sounds are pering a dozen or more intervals whereas a ceived may be expected to differ considerably tone language of just a few contrastive levels from an instrumental analysis of the same may seem extraordinarily difficult to them group of sounds Modern theories of percepPerhaps the explanation for this lies in the tion postulate an active decision-emitting fact that speech intonation such as is emlistener who busily constructs a perceptual ployed in a non-tonal language like songs model of his world by selectively abstracting merely add extra shades of meaning to lexical information from the stimuli surrounding him meanings Tonal languages in contrast to Any attempt on our part to process speech by this employ pitch as an intrinsic part of machine must take into account this propensity each word on the part of the human mind The words that The use of pitch differences to convey are used in speech communication are of prilinguistic information appears to be unimary importance of course but often the versal and many languages share the same or real message is contained in how they are similar practices In nearly all languages said Somehow we must discover how to simua falling pitch signifies the completion of late human perception in an area which is a grammatical unit such as a normal sentence difficult to describe with precision and is On the other hand an incomplete utterance one even more resistant to understanding Frequency in Hz 1 L Complex Tones May 79 CRYPTOLOG Page 15 P L UNCLASSIFIED 86-36 DOCID 4009827 UNCLASSIFIED NSA-Crostic No 25 By D H W For' the benefit of readers who Jomplain that these puzzles ax e too easy this one is given in diagramless fo 1'fT1 Reader'S who do not feel this way may find r'e lief on page 12 DEFINITIONS WORDS A How to express the most well-known line of the Rubaiyat in telegraphese B Sweet dish C 11 even by God Browning's Pax a JelsuB 5 wds D Humphrey Bogart's only line in his Broadway debut-really 2 wds E Having virtually unlimited authority F Civil _ G Why explained the circus owner to the departing human cannonball a replacement was going to be difficult to locate 8 wds H Openings I Summer outdoors machinery J Low-priority telegram 2 wds K Archie Bunker's daughter's telegram explaining that her Sunday arrival would be delayed 24 hours because of a bus breakdown 4 wds L 44 251 23 ---g j 217 267 182 235 167 12 116 137 Try M and we' 11 tell you what you are afraid of Robert Benchley 4 wds 232 173 183 N 7 52 3 wds O Princess of Monaco's roadster P A people of Asia Minor 2nd millenium B C Q Caesar considered this to be nothing R Powerless ineffectual lacking vigor May 79 CRYPTOLOG Page 16 UNCLASSIFIED 8T 92 --------- -- ---- 1 DOCID 4009827 UNCLASSIFIED S Seat less leather trousers T One who bears witness U Fictional land of little people V Planet W Elevator man X Highest in position place or rank Y Latin-American dance var Z North by a Rooflike cover over a window b Flat even smooth Z3 G4 T 5 o6 C7 L8 J 9 G 10 Jll M 12 It 13 N 14 H 15 o 16 D 17 C A 19 X 20 It 21 o 22 U 23 It 24 N 25 J 26 o 27 M 28 H 29 G 30 C 31 A 32 G 33 2 34 0 35 L 36 J 37 G 38 X 39 L 40 X 41 G 42 I 43 E 44 It 45 L 46 M 47 H 48 B 49 R 50 G 51 L 52 P 53 D 54 B 55 N 56 X 57 R 58 R 59 E 60 J 61 F 62 G 63 W64 F 65 66 X 67 G 68 E 69 M 70 a 171 G 72 C 173 Ul 74 P 15 N176 G77 KI78 A 79 o 80 D 81 M 82 C 83 J 84 G 85 C 86 I 87 S 188 0 89 W 90 P 191 P 92 M 93 a 94 Z 95 U 96 b 97 It 98 R 9 1 o 18 2 103 A 104 a 105 B 106 a 107 E 108 V 109 Z 110 X 111 I I G 100 I 101 0 102 W 112 Y 113P 114 L 115 I 1161 117M 118L 119 A 120 G 121 W 122 B 123 b 124 X 125 G 126 a 127K 128 F 129 I 130 I 131 G 132 A 133 F 134 N 135 C 136 Z 137 K 138G 139 C 140 J 141 A 142 D 143 N 144 H 145 0 146 T 147 'I 148 L 149 H 150 F 151 U 152 Z 153 Q 154 U 155 b 156 a 157 G 158 R 159 C 160 E 161 A 62 G 163 R 164 II 165 H 166 N 67 171 C 172 G 173 M 174 I 175 T 176 A 177 b 178 T 179 R 180 C 181 H 182 I I 168 E 169 b 170 0 183 M 84 G 185 G 186 V 187 M 188 Y 189 I 190 0 191 C 192 K 193 G 194 0 195 J 196 C 197 'I 198 G 199 N 200 5 201 K 202 1 203 B 204 E 205 A 206 G 207 Y 208 V 209 0 210 I 211 P 212 H 213 M 214 G 215 E 216 U 217 222 I I 218 J 219 Q 220 G 221 T 223 T 224 E 2255 226 Q 227 B 228 U 229 C 230 G 231 E 232 M 233 B 234 M 235 K 236 N 237 Q 238 R 239 A 240 G 241 V 242 J 243 K 244 F 245 Z 246 0 247 V 248 G 249 C 250 G 251 256 V Z57 M Z58 Q 259 I I 260 Z 261 T 1262 S 263 0 1264 A 265 C 2M M 267 KIZ6tl N 273 X 274 P 275 N 276 J 277 T 278 2 279 0 2tlO C 1281 X 282 V 2tl3 u Ztl4 my 79 CRYPTOLOG Page 17 UNCLASSIFIED U 252 M 253 A 254 D 255 0 Z6 Y UUIi IZIl U ZlZt' IZtl5 M Ztl6 IZtll H 'z8tl I dhw DOClD 4009827 UNCLASSIFIED Continuing with the observations of Sydney Fairbanks on CWTent American English CRYPTOLOG offers his remarks from the July 1958 issue of The NSA Technical Journal covering among other TOCltte1'S i1'1'egardless lJ1' Fairbanks' cormrents are just as timely now as on the day they were first published more Fai banR$ oon lJ1' Englith wo different garage mechanics have assured us that the starter on the car is suffering from a defect in the cellunoid After due inquiry we incline to the belief that the word is the offspring of celluloid and solenoid With the advent of plastics celluloid must be disappearing rapidly from the popular vocabulary and it is pleasant to think that it leaves one descendant however illegitimate If the object in question were a solenoid made out of celluloid--but the engineers tell me this is improbable--we should have a case of a portmanteau word generated as it were spontaneously whereas most such words are deliberate acts of creation Lewis Carroll who christened them exploited the idea more fully than anyone else--except of course James Joyce--and the preface to The Hunting of the Snark is recommended reading for those who plan to embark with Ulysses But the whole point of Carroll's inventions such as frumious for fuming-furious is that they are self-conscious In only one case- chortle presumably from snort and chuckle --has the creature sneaked through the barrier of literature and moved into the outside world where Jabberwocky is no longer a part of the scenery The same may be said of like inventions of humbler origin Aggranoying of obvious parentage enjoyed a somewhat dreary vogue in England shortly after the death of the Great White Queen being used chiefly in discourses addressed to the very young It was in a class with thusly which was likewise intended to be recognizedO as an amusing blunder But we cannot remember that irregardless by irrespective out of regardless was ever a humorous invention although Webster Er1'On 01' humorous U S gives it the benefit of the doubt In our editorial capacity alas we meet with both thusly and irregardless employed totally without humorous intent The word insinuendo offered by a friend of ours seems to us full of charm and deserving of an appreciative audience yet the thought of having it pass into the irregardless group acts as a serious deterrent Better perhaps to strangle V it at birth What chiefly daunts the editor however in that endless hopeless rearguard action against neologism to which all his kind are dedicated is not the hybrid word but the hybrid construction Termed incorrect for instance and described as incorrect meet each other in some ninth circle of the popular mind and termed as incorrect emerges He said he was going and He stated that he was going give He stated he was going or even Goodbye he stated I convinced him that he should go and I persuaded him to go produce I convinced him to go I saw him going downstairs and I was told of his going upstairs combine in I was told of him going downstairs Other than and different from beget different than Much alike and very different yield much different As regards and in regard to create in regards to Equals and is equal to give rise to equals to And so on from here to Mesopotamy The general principle that if two words are equivalent in any respect they must be equivalent in all respects moves on remorselessly and the world has grown grey at its breath In such moods we try to take comfort in the historic approach Dean Swift felt with perfect justice that mob a slang abbreviation of a stale classicism mobile vulgus was the sort of word that no one with any sense of style would permit himself to use but the earth continued in its orbit the equinoxes no doubt precessed and mob became one of our better monosyllables The whole English language if it comes to that is an undignified sloughing off of fine old Germanic terminations made possible by the adoption of a monotonous standard word order in place of the ancient freedoms And even before that if good little Indo-Europeans had listened more carefully to the starred forms used by their mothers there would have been no vulgar sound-shift and we should all be talking a pure and original tongue If good little apes o but by now we have had all the comfort we can take at one time May 79 CRYPTOLOG Page 18 UNCLASSIFIED DOCID SECRE'f SPOKE 4009827 I IT123 OHAPENKO SHAPENKO-What Difference Does It Make May 79 CRYPTOLOG P L u Page 19 BECRET SPOKE iii EO 1 4 c P L 86- 3 6 86-36 DOCID SECKE' ' SPOI E 4009827 News of the Communications Analysis Association u CAA BOARD FOR 1979 U u President President-Elect Treasurer Secretary At-large At-large At-large Programs Crypto-TA SIG Cryptologic History SIG 0 36715 33695 49835 55855 49355 54935 59065 0 44525 44665 Dan Buckley 21715 Board meetings are held the last Wednesday of the month at 1000 hours in the Al Conference Room 2Nl09 Have you joined the CAA If you would like to do so send one dollar and an application form you can get one from any board member to the Treasurer LINGUITRIVIA U o U In a recent issue of Russkaya Rech' Russian Speech the author of an article on the orthographies of the various languages of the Soviet Union states that 00 of the 66 written languages of the USSR have writing systems based on the Russian alphabet He does not identify the remaining six U Five of them come to mind immediately Armenian Georgian andthelanguagesofthe three Baltic republics-Estonian Lithuanian and Latvian But what is that sixth one U Does any reader know the answer If you do please let CRYPTOLOG know May 79 CRYPTOLOG Page 20 BECKE' ' SPOKE EO 1 4 c P L 86-36 P L 86-36 P L 86-36 DOCID 4009827 UNCLASSIFIED CLUB AND LEARNED SOCIETY OFFICERSl The pages of CRYPTOLOG are open to you for announcements concerning the activities of your organizations-meetings banquets lectures election results etc CRYPTOLOG comes out on the first of every month Submissions should be turned in six weeks prior to the publication date For further information call 3957s OLD SUBSCRIBERS NEW SUBSCRIBERS To change your address on an existing subscription or to begin a new one call 3957s or drop a note to PI CRYPTOLOG May 79 CRYPTOLOG Page 21 Pi-Apr 79 S3-705 UNCLASSIFIED DOC 827 ------ 'fillS B6etJI'tIEN'I' e6N'I'AiNS e6BEWORB PfIA'I'ERIAL 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