HAS GERTRUDE STEIN A SECRET?

BY B. F. SKINNER I

lN the Autnbi,ography of Akce B. ToklasGertrude Stein tells in the followingway of some psychological experimentsmade by her at Harvard: -

She was one of a group of llarvard menand Radclifre leomen and they all lived veryelosely and very interestingly together.O:le of them, a young philosopher andmatlematician who was doing researchwork in psychology, left a deff"ite mark onher life. She and he togetier worked out aseries of experiments in automatic writingunder tle direction of Mi.i-nsterberg. Theresult of her own experimeats, which Ger-trude Stein wrote down and which wasprinted in the .Earrard Psych&giml Re-nerc, was tle first *"iti"g of hers ever to beprinted. ft is ser.v int$esting to reod be-cause the mettrod of writing to be after-wards develo@ ta Three Li'oes and TheMakina oJ Amerimrw already shows itself.

There is a great deal more in thisearly paper than Miss Stein points out.ft is, as she says, an anticipation of theprose style of Three Liaes and is un-mistakably the work of Gertrude Steinin spite of the conventional subjectmatter with which it deals. Manyturns of speech, often commonplace,which she has since then in some subtleway made her own are already to befound. But there is much more thanthis. The paper is concerned with anearly interest of Miss Stein's that musthave been very important in her laterdevelopment, and the work that itdescribes cannot reasonably be over-

50

r

Iooked by anyone trying to understandthis remarkable person.

Since the paper is hard to obtain,Ishall summarize it briefly. It was published in the Psychol,ogical Retiew forSeptember 1896 under the title, 'Nor-mal Motor Automatism,' by Leon M.Solomons and Gertrude Stein, and itattempted to show to what extent theelements of a 'second personality' (ofthe sort to be observed in certain casesof hysteria) were to be found in a nor-mal being. In their experiments theauthors investigated the limits of theirown normal motor automatism; that isto say, they undertook to see how farthey could 'split' their own personali-ties in a deliberate and purely artificialway. They were successful to the ex-tent of being able to perform many acts(such as writing or reading aloud) in anautomatic manner, while carrying on atthe same time some other activity suchas reading an interesting story.

il

In the experiments with automaticwriting, a planchette of the ouija boardtype was originally used, but as soon asthe authors had satisfied themselvesthat spontaneous writing movementsdo occur while the attention is directedelsewhere, an ordinary pencil and paperwere used instead. The subject usuallybegan by making voluntary randomwriting movements or by writing the

HAS GERTRUDE STEIN A SECRET? 51

letter m repeatedly. In one experimentthis was done while the subject read an

interesting story at the same time, andit was found that some of the wordsread in the story would be writtendown in an automatic waY. At firstthere was a strong tendency to noticethis as soon as it had begun to happen

and to stop it, but eventually the wordscould be written down unconsciouslyas well as involuntarily. (I shall use

Miss Stein's psychological terminologythroughout.)'sometimes the writingof the word was comPletelY uncon-scious, but more often the subject knewwhat was going on. His knowledge,however, was obtained by sensations

from the arm. He was conscious thathe just had written a word, not thathe was about to do so.'

In other experiments the subjectread an interesting story as before, andsingle words were dictated to him to be

written down at the same time. These

were difficult experiments, but afterconsiderable practice they were suc-

cessful. The subject was eventuallyable to write down 'five or six' wordsspoken by another person' without be-

ing conscious ofeither the heard sounds

oith" movement of the arm. If his

attention were not sufficiently welldistracted he might become aware thathis hand was writing something. Theinformation came from the arm' notfrom the sound of the dictated word.'It is never the sound that recalls us.

This, of course, maY be an individ-ual peculiarity to a certain extent.Yet, Miss Stein has a strong audi-tory consciousness' and sounds usu-

ally determine the direction of herattention.'

In a third group of exPeriments thesubject read aloud, preferably from-anuninteresting story, while being read tofrom an interesting one' 'If he does notgo insane during the first few trials, he

iiU qui"t ty learn to concentrate his

attention fully on what is being read tohim, yet go on reading just the same.

The reading becomes completely un-conscious for periods of as much as apage.' Automatis reading of this sort isprobably part of the experience ofeveryone.

The fourth and last grouP bringsout the relevance of the erperimentsto the later work of Gertrude Stein.I shall let Miss Stein describe theresult.

Spontaneous automa'ti.c wriling. - Thisbecame quite easy after a little practice.We had now gained so much control overour habits of attention that distraction byreading was almost unnec€ssar.v. lfiqsStein found it sufrcient distraction often tosimply read what her arm wrote, but fol-lowing tlree or four words behind her

pencil. . . .- A phrase would seem to get into tie head

and keep repeating itseH at every oppor-

tunity, and hang over from day to da5 esT'The stuff written was gremmatical, and tlewords and phrases fitted together ail righl'but there was not much comected tloughrThe unconsciousness was broken into ever.s

six or seven words by flashes of conscious-

ness, so tbatone cannot be sure butwhat tleslight element of conaected ttrought rhichocLsionally appeared was due to t}<=e

flashes of consciousness. But tle ability towrite stuff that sounds all right" witloutconsciousness, was fairly well demonstratedby the experiments. Here are a few qreci-

mens:'Hence there is no possible way of asoid-

ing what I have spoken of, and if rhis is not

bJieved by the people of whom you harespoken, then it is not possible to prevent

tle people of whom you have spokea so

glibly. . . .'" He"e is a bit more poetical thrn in-

telligible:'frhen he could not be the longest and

thus to be, and thrx to be, tle strongest"A:rd here is one ttret is neither:

'This long time whea he did this bes-i

time, and he could tlus bave been bound'

and in thi* long time, when he could be tlisto first use of ttris long time. . . .'

RO THE ATLANTIC MONTHLY

III

Elere is obr-iousll- an important docu-ment. No one s-ho haq rex.,d TenilerBultans or the later work in the samevein can fail to recognize a familiar notein these examples of automatic writing.They are quite genuinely in the mannerthat has so co--onl1' been taken ascharacteristic of Gertrude Stein. MissStein's description of her experimentalresult is eractlJ- rhrt of the averagereader confronted with Tm.d,er Buttonsfor the first time: 'The stuff is gram-matical, and the words and phrases fittogether all right but there is not muchconnected thought.' In short, the caseis so good, simpl-v on the grounds ofstyle, that yie are brouglt to the s-lyiftconclusion thql the rwo pro'ducts havea common orfuin, and that the w'ork ofGertrude Stein in the Tender Buttonsmanner is written autorruxticall)' andunconsciously in some sr.lch niav as thatdescribed in this earll' paper.

This eonclusion grows more plausibleas we consider the ca-se. It is necesary,of enurse, to disti-ogirish betw€en theGertrude Stein of Three Litw and tieAutnbingraphy and the Germrde Steinof Tender Butlons, a distincti,on that isfairly easily made, eren tlougfr, as weshall see in a moment" tbere is some ofthe first Gertrude Stein in the htterwork. If we confine ourselve for thepresent to the second of these tyio per-sons, it is clear tbat th€ hlpothetielauthor who might be inferred from thewriting itself possesses just those char-asteristics that we should erpect to f,ndif a theory of automatis q-riting werethe right answer. Thus there Lc r-ery-little intellectual content discoverable.The reader - the ordinary reader. atleast - cannot infer from the writingthat its author possesses any consistentpoint of view. There is seldom any in-telligible expression of opinion, andthere are enough capricious reversals to

destroy the effect of whatever theremay be. There are even fewer emo-tional prejudices. The writing is cold.Strong phrases are almost wholly lack-ing, and it is so difficult to find a well-rounded emotional complex that if oneis found it may as easily be attributedto the ingenuity of the seeker. Simi-larly, our hypothetical author shows nosign of a personal history or of a cul-tural background; Tender Buttons isthestream of consciousness of a womanwithout a past. The writing springsfrom no literary sources. In contrastwith the work of Joyce, to whom asuperficial resemblance may be found,the borrowed phrase is practicallylacking.

When memorized passages occur,they are humdrum - old saws orsimple doggerel recovered from child-hood and often very loosely para-phrased: 'If at first you don't succeedtry try again,' or 'Please pale hot,please cover rose, please acre in thered. . . . ' If there is any character inthe writing whatsoever, it is due to thissavor of the schoolroom, and the oneinference about the author that doesseem plausible is that she has been tograrnmar school. Her sentences areoften cast as definitions ('What is aspectacle a spectacle is the resem-blance . . 'or'A sign is thespecimenspoken') or as copy-book aphorisms('An ercuse is not dreariness, a singleplate is not butter,' or 'There is coagu-lation in cold and there is none in pru-dence') or as grammatical paradigms{'I begin you begin we begin they be-girn se began you began I began').This heary dose of grammar school isespecially strongly felt in An Eluciila-frbn, Viss Stein's first attempt to ex-plain [6mslf, and a piece of writing inwhich tiere are many evidences of astruggle on the part of the consciousGertrude Stein to accept the origin ofthe Tender Buttorw manner. Miss Stein

HAS GERTRUDE

wanted the volume Lucy Church Ami-ably to be bound like a schoolbook, butI shall leave it to a more imaginativemind to elaborate this metaphor fur-ther.

This is apparently as much of thewriting as will help to illuminate thecharacter of the writer. For the rest, itis what Miss Stein describes as sound-ing all right without making sense.There is no pa,radox about this, there isno secret about how it is done; but itgives us very little information aboutthe author. Grammar is ever present

- that is the main thing. We are presented with sentences ('sentences andalways sentences'), but we often recog-nize them as such only because theyshow an accepted order ofarticle, substantive, verb, split infinitive, article,substantive, connective, and so on.The framework of a sentence is ttrere,but the words tacked upon it are an oddcompany. In the simplest type of casewe have a nearly intelligible sentencemodified by the substitution for a singleword of one sounding much the same.This sort of substitution was report-ed by Miss Stein in connection withher experiments in automatic reading:'Absurd mistakes are occasionally madein the reading of words - substitutionssimilar in sound but utterly different insense.' The reader will recognizn it asthe sort of slip that is made when one isvery tired. In more complex cases itcannot, of course, be shown that theunintelligibility is due to substitution;if most of the words are replaced, wehave nothing to show that a word is aslip. We must be content to charac-tet'rze it, as Miss Stein herself hasdone: 'We have made excess return torambling.'

IV

From this brief analysis it is apparentthat, although it is quite plausible thatthe work is due to a second personality

STEIN A SECRET? 53

successfully split off from Miss Stein'sconscious relf, it is a very flimsy sortof personality indeed. It is intellectu-ally unopinionated, is emotionally cold,and hes no past. It is unread and un-learned beynnd grammar school. ft isas easily influenced as a child; a heardword *ay force itself into whateversentence may be under construction atthe moment, or it may break the sen-tence up altogether and irremediably.Its literary materials are the sensoryrh;ngs n€arest at hand - objects, sounds,taste, smells, and so on. The readeruray compare, for the sake of the strongcontrast, the materials of 'Melanctha'ia Three Lhtes, a piece of writing ofquite another sort. In her experimentalwork it was Miss Stein's intention toar-oid the production of a true secondpersonality, and she considered herselfto be successful. The automatism shewas able to demonstrate possessed the'elements' of a second personality, itwas able to do anything that a secondpersonality could do, but it never be-came the organized aker ego of thehysteric. The superficial character ofthe iaferential author of Tender Buttonsconsequently adds credibility to thetl"ory of automatic authorship.

The Gertrude Stein enthusiast mayfeel that I am being cruelly unjust inthis estimate. I admit that there arepassages in Tender Bultons that eludethe foregoing analysis. But it must bemade clear that the two GertrudeSteins we are considering are not keptapart by the covers of books. There is agood deal of the Gertrude Stein of theAutnbio gr aphg in T end.er Buttons, in theform of relatively intelligible comment,often parenthetical in spirit. Thus atthe end of the section on Mutton(which begins 'A letter which canwither, a learning which can suffer andan outrage which is simultaneous isprincipal') comes this sentence: 'Ameal in mutton mutton why is lamb

54 THE ATLANTIC MONTHLY

cheaper, it is cheaper because so little ismore,' which is easilY recognized as afavorite prejudice of the Gertrude Steinof the Atnobi.ography. Similarly such

a pbrase as 'the sad procession of theunkilled bull,' in An Elucidat'ion, isplainly a reference to another of MissSt"io's interests. But, far from damag-ing our theory, this occasional appear-anee of Miss Stein herself is precisely

what the theory demands. In her paper

in the Psychological Reaiew she deals atlength with the inevitable alternationof conscious and automatic selves, andin the quotation we have given it willbe remembered that she commentsupon these 'flashes of consciousness.'

Even though the greater part of Tander

Buttons is automatic, we should expectan 'element of connected thought,' andour only problem is that which MissStein herself has considered - namely,are we to attribute to consciousflashes all the connected thought thatis present?

There is a certain logical difficultyhere. It may be argued that, since wedispense with all the intelligible sen-

tences by calling them conscious flashes,

we should not be surprised to find thatwhat is left is thin and meaningless.IVe must therefore restate our theory,in a way that will ar-oid this criticism.TVe first diride tle rritings of GertrudeStein into two parts on the basis oftheir ordinary intelligibilitl-. I do notcontend that this is a hard and fast line,but it is a sufrcientll- real one for mostp€rsons. It does not, it is to be under-itood, foilo*- the outline of her worksIVe then show tbat tle unintelligiblepart has the characteristics of the auto-matic writing produced b.v lliss Steinin her early psychological erperiments'and from tlis and ttrany other con-siderations we conclude that our divi.sion of the work into two parts is realand valid and that one Part is auto-matic in nature.

I cannot find anything in the Auto-biographg or the other works I haveread that will stand against this inter-pretation. On the contrary, there aremany bits of evidence, none of whichwould be very convincing in itself, thatsupport it. Thus (1) Tender Buttonswas written on scnrps of pa.per, and noscxap was ever thrown away; (2) MissStein likes to write in the presence ofdistracting noises; (3) her handwritingis often more legible to Miss Toklasthan to herself (that is, her writing is

'cold' as soon as it is produced); and(a) she is 'fond of writing the letter raz,'

with which, the reader will recall, theautomatic procedure often began. InAn Ehtci.datipn, her 'first effort toreafize clearly just what her writingmeant and why it was as it was,' thereare many fitful allusions to the ex-perimental days: 'Do you all under-stand extraneous memory,' 'In thisway my researches are easily read,' asuddenly interpolated 'I stopped Istopped myself,' which recalls themajor difficulty in her experiments, andso on.

v

It is necessary to assume that whenGertrude Stein returned to the practiceof automatic writing (about 1912?) shehad forgotten or was shortly to forgetits origins. I accept as made in per-fectly good faith the statement in theAutobiography that'Gertrude Steinnever had zubconscious reactions, norwas she a successful subject for auto-matic sriting,' even though the evi-dence to the contrary in her early paperis incontrovertible. She has forgottenit" just as she forgot her first novelalrnost immsdislsly after it was com-pleted and did not remember it againfor twenty-five years. It is quitepossible, moreover, that the manner inwhich she writes the Teniler Buttonssorl of thing is not unusual enough to

I{AS GERTRUDE STEIN A SECRET? 55

remind her of its origins or to be re- come to tnow Picasso and Matisse and

;;;il'tv "trr"ir- 6rr" of the mosr was already long in the practice of de-

interesting statements ir "tft"

".q.na f=ai"S their woik against the question'

ouoted from her earlv prp". i.1n"t c.r- 'Ifbat do€s it mean?' with such an ex-

l,uj; ##'i;;"d'i Jr{ficie't distrac- perience behind one, it is not difficult to

tion simply to follow what she was accept as art what one has hitherto

writing some few *;tir;hi"d her di*ii*"d as the interesting and rather

oencil. rf in the "ouir"

of time she surprising result of an experiment. rtfj. "fr*" ilrr"* rr"i attention n@Fer *.i t 6ri"nu, -only

because Gertrude

and nearer to the p"""ii,- rrr" *ro stein had alredy prepared the defense

eventually frurr" ,"u"tt"J * poio-t "t as it applied to Picasio that she could

which there remained only the finest put foJh her own unintelligible. prod-

distinction between 'knowing what one uct as a serious artistic experirnent'

is going to write ""d;;;*it? 'r91try fgitgtryo"of thesoundirrtelligenceof

has written it.' Ttis iJ; tr?;t,t"r"l u-* stein there is a great natural re-

state to which Miss stein devotes on- sistance against the production of non-

siderable space in her paper' It is snse' It ias the *ijot nroblem in her

therefore reasonable for us to assume "-p.ri-".rttt techniquei'I stopped i

that the artificial character of the ex- sopped.myself.' But the writing suc-

perimental pro*ao*'},u, completel;- T*:t in this case' because the re-

i;;;;ii;;rh;;;h;L"i"''""rr-l sistanc€ had been broken down' first

not-far-from-normal "t"t" i"t" n-hich by the procedure,of the experiments'

Miss Stein *oy pu,, *o,,'"'p*,irgr,-

*hicn.pe'*itted the sustained produc-

enough and in *uicn-tii fide' i'it- ti'onof meaningless sentences' andlater

lons style i* fortn"o*i,'g'" '

^ - . . Ui-1n" champloning of Picasso' which

Having begunto pioali""stuff ofthl. permitted their publication. This was

sort again, howe1er,'i\{i" St"io "oUa a fortunate combination of circum-

not have failed to notice its peculiari *""" 'I could explain" she says in

ties. wehaveher"*;;;;;;;il.b" i Elurid,otion, 'how it happened

;;";;.;";;; il;i; a111qmadc accidentally that fortunatelv no ex-

writing do not ,fto*-L""f' connected planation rvas necessary''

thought,andIb"li".,"*"arefully.}Ii."steinhasnot,however'freediustified in our

"turo""t"ri^tio" of thu herself from the p-roblem of the mean-

'-Ht"'r" ;il"'.;"?;;irr--n-iuo"t u ing of the things ihe writes' She is not

;ordinarily unintelligible.' I know'hat a6"e being Uotnqlg{.by criticism on

it would be quite possible for an in- th" ,"or" of"unintelligibility' she often

dustrious and irrgeniJil-p".,* 1o n.a characterizes her work in this vein as

anv number or **nir!", irl it,lG * i, eroerimental, but that is in no sense an

iJ"JJJt'il;'; ;il"d*r,i.g, io 1or- exolanarion. Bevond this her answer

chance *rrorrg"-"rri'.i--*3ia". But #ms to be that the writing is its orvn

it " "o.r"ttrion to which we are nos justification'

led is that the *o,k- *ith which we It was not a question' she told

are dealing i, .r"rf pr"u"liv ""r"- her-o*ford audienc-e, of whether she

telligible in any o'di"oty sense' not "* iigttt in doing the kind of writ-

only to o,tu, ,"uiJrr,'d";';}i"o, i"g rrrJaia. 'She had been doing as she

stein herself. wr,v,"'trr"", did sbe dii fot about twenty I'ears and now

publish? tllr*trrt"d to hear ie"r lecture'' And

It is important lbr our theory that tft"" ftua qre{gus]y .dealt rvith the

between 1g96 and rirrg Mi., stein had matter in an Elucidation:-

56 THE ATLANTIC MOI{THLY

II it is an event just by itself is there aquestion

Tulips is there a questionPets is there a questionFurs is there a guestionFolds is there a questionIs there an)"thing in question.

I think we must ac€ept this answer tothe ethical question of whether she isdoing right by Oxford and the King'sEnglish. The final test of whether it isright is whether anyone likes it. Buta literary composition is not 'an eventjust by itself,' and the answer to MissStein's que{v- is that there certainly arequestions, of a critical sort, that maylegitimately b€ rais€d. Ileaning is oneof them.

One kind of meaning that might befound if our theon' is ralid is psycho-logical. In noting the presence of ver-bal slips ('zubstitutions similar insound but utterl;,- difierent in sense') welay ourselves op€n to the criticism ofthe Freudian, who would argue thatthere are no true slips. According tothis view, there is alwals some reasonwhy the substitution is mads, and thesubstituted r*'ord will har-e a deepersignificance if q-e can find it. But weare here not primarilv cpncemed withsuch psychological signffi ca nces.

Of literary significances it may beurged that for the initiated or sJrrn-pathetic reader there is an intellectualcontent in this part of \fiss Stein'swork that rve have or-erlooked. \ow,either this w-ill be of such a sort that itcould also be erpressed norrnally, or itwill be a special kind of content thatrequires the form given to it by lIissStein. A partisan could so easil1- provethe first case by transhting a r€phesen-tative pa,ssage that we me]'assume itnot to be true. The second cas€ re-quires a very difficult theor-r. of howl-edge in its defense, and we shall notneed to inquire into it any more closely.It is quite true that something bapp"*

to the conscientious reader of TenilerButtons. Part of the effect is certainlydue either to repetition or to surprise.These are reeognized literary devices,and it may be argued that still a thirdkind of meaning, which we may des-ignate as emotional, is therefore to befound. But in ordinary practice thesedevices are supplementary to expres-sions of another sort. The mere genera-tion of the effects of repetition andsurprise is not in itself a literaryachievement.

VI

We have allowed for the presence ofany or all of these kinds of meaning byspeaking only of ordinary intelligibiLity. I do not think that a case can bemade out for any one of them that isnot obviously the invention of theanalyzer. fn any event the presentargument is simply that the evidencehere offered in support of a theory ofautomatic writing makes it more prob-able that meanings are not present,and that we need not bother to look forthem. A theory of automatic writingdoes not, ofcourse, necessarily excludemeanings. It is possible to set up asecond personality that will possess allthe attributes of a conscious self andwhose writings will be equally meaning-ful. But in the present case it is clearthat, as Miss Stein originally intended,a true second personality does not exist.This part of her work is, as she hascharacterized her experimental result,little more than 'what her arm wrote.'And it is an arm that has very little tosay. This is, I believe, the main impor-tance of the present theory for literarycriticism. ft enables one to assign anorigin to the unintelligible part of Ger-trude Stein that puts one at ease aboutits meanings.

There are certain aspects of prosewriting, such as rhythm, that are notparticularly dependent upon intelligi-

HAS GERTRLTDE

bility. It is possible to experiment n-iththem with meaningless words, and itmay be argued that this is n-hat ishaplrcning in the prcsent case. Con_sidering the freedorn that Nliss Steinhas given herself, f do not think theresult is very striking, although this Lclearly a debatable point. It I a fairerrn{.erpretation, however, to suppose. inaccordance with our theory, that thereis no experimentation at ihe time thewriting is produced. There mar- begood reason for publishing the maierialafterward as an experiment. Forexample, I_recognize the possibilitl- of asalutary, though accidental, effect"upooGertrude Stein's conscious p*' o.upon English prose in general. InCompotition As Eaplanatibn, fo,

"r_ample, there is an intimate fusion ol t.hetwo styles, and the conscious pa_{sag€sare imitative of the automatic srvle.This is also probably true of pan-. oithe Autobiography. It is perbaps im_possible to tell at present s.herher theeffect upon her conscious prose is an.r _

thing more than a loss of discipline. Tbcompensating gain is often r:eqr- _:reat-

S?EI\ A SECRET? 57

We har-e no reason, of course, toestimate the literary value of this nartof IILqs Stein's work. f t might be con_siderable, even if our theor/is correct.It is apparent that Miss Stein believesit to be important and has accordinslvpubllhed it. If she is right, if this pirtof her x-ork is to become historically assigal{cant as she has contended, ihent.he importance of the document withrxhich we began is enormous. For thefrrst time we should then have anacoount by the author herself of howa literary second personality has beenser up.

I do not believe this importancee-tist.<, horr-ever, because I do not be_

F:-" g the importance of the part of-\Iiss Stein's writing that does noi mrkes€*se. On the contrary, I regret theunfortunate effect it has had iriobscur-ing the finer work of a very fine mind.I x-elcome the present thelry becauseir gir-es one the freedom to dismiss onepan of Gertrude Stein's writing as apr-obably ill-advised experiment and toenjo.v the other and very great part*ithout puzzlement,