archiving

: con-

I~rrangement: I Levels of Control

Archival arrangement involves both ensuring that a Gody of records ia in a meaningful internal order and placing the records within the overall system according to which the repository's holdings are oganizdl. The distinction betwen t h w two related operations iB important. Equally important is the differenm between physical and intellectual ar- r a n g e m e n ~ e e n . arranging actual box= and manipulating information about records. With these distinctions in mind we can apply our definition of arrangement as the pr- of organizing and man- aging historical records by:

1) identifying ox bringing together sete of records derived from a common source which have common chamcteristica and a

group/manu&pt collection, seriea, file unit, and document. The claseric exposition of this approach came in a 1964 American Archivist article by Oliver W. Ho1rnes.l This way of prmeeding was driven by the perceived requirement on the part of archivim to assign every document to one file unit (such ag a folder), every Fde unit to one series and wery series to one ofice of origin. This system for physical control and arrangement proved easy to apply. Using it for documents in two of our sample record sets would produce the following:

(Repitory1 State Arthives University Library 1) Repository Public Records Regional History

Division Center common file structure, and 2) Rec. Grp.! Dept. of Nat. h. Southside Gamrn. 2) identifying relationship among seta of Cull. Sew. records and between records and their cre- 3) Series Park Planning House Matron's

atom. Files Monthly &posts

W definition can be applied to m y set of rec- 4) File Unit Capital County Reports, 191b15 or& that is to be arranged, from the m r d s of an (folder title) Park Plans. entire organhzation to a few boxes from a small pra- 197678

within an agency. 5) Document 1973 Middletam January 1911 Municipal Park Report

General Approach to Arrangement Plan

The operational approach which has domi- The difference between archives and manu- nakd arrangement in the United States is straight- scripts in cases like these derives from the way rec- forward and practical. It emanates mainly from the ords are acquired. The actual work of physical ar- experience of the National Archives in controlling rangement in axchives usually kgina at the series r n m of rncdem secords. Its core is the concept of level, because that is the set of records most com- a five-level hierarchy of arrangement moving from the larger scale t6 the smder-repository, record Wolmes, "Archival Arrangement," 2241.

60 Arranging and Wrib'i Archim and Manuscripts

monly q u i d . Rather than receiving all the mc- o r d s o f t b D N R ~ r ~ o f t h e O f I h o f PmkPlplan- nhg, the data archiva ia far more h l y on a daily basis to e v e part of some wria of f11es such as the planning records. The record p u p , such RS the DNFi, which ~erpes as s general admhbhtive con- struct, has ahady h n estabW4 though of coursa new record groups are established when m e ords are m f v e d from agencies not previoul31y rep raented ia the d v e a In contrast to the usual d v a l concentration on serk, m a n d p t a a s merit dads mom ofbn with en& o o U & ~ , such as the southside Community SerPicea fec0rd3, -use in m d p t ~poaitories t h w are the more @iid of d acquired.

The fivepart r n d was m n modified to fit actud practice by the addition of two &her "lev- eb"+ubgmups within reoord graupe/cullectionn and 8 ~ - witbin mries. They may be defied ~ E I

follm . S u ~ u p s are Idk of r%wrcte mahlsrinnrl o r c m a t e d b y a ~ ~ t e ~ body, or group of aeriee dated by a-n activity or UBB.

* - a r e d w - j n & ~ d W l a r g e r A l e ~

The numbr of sbgm d subgroup and mhc+ riee can in theory be extended endledy. There can be dgmups of subgroup in complex organizations, and mabee* of eubseriee ia complex f u strue tureer. However, archi* should try to nninimize the complexity of axrangemen& so that both lwela &odd b Iimikl if pssible to a few h g t s But the umfulnwa of tbe additionai levels is apparent if we return b our model obilectio~. The DNEE rew+ would be mu& mom undmitmdable if the whole agency structure could be displayed..

Record Group: Depmbmt of Natural R e 80-

S-up Bumau of Parb and -tion S u k u m p : MICB of Park PZElnning

The example of the Swthde Community Ser- vice~ recoda is perhap more hportant. At the time the Monthly Howe Matron's mprta were mated them was no Southside CommuniQ Sewices. but in- daad the Southside Settlement The o r g a n h a m origin of the mportg is h i t repmnt%d by the following:

Colldon: Southside Community Services Subgroup: Souhide W e m e n t House 119C%1948) Seriee: H o w matron'^ Monthly fiepork I

In terms of ~ubewies, the full file structure within the 0E.w of Park Pl- leading to the 197&?8 Capital a u t y plans b I

Serim: Park Planning Files S-88: County Parb . s u u r l m : Cepital County P I m File Unit: Capital Couotp, 1976-1978

T h i ~ full system emp- organhition in tern of the provenance and original order of the records. It a h helps iuwtrate what amhivista cal l the "evidential value" of mrds. Evidentid d u e &em to the way the -tion and file st- of records revads hiformation about the functions and activitiw of an mganhth. However, this way of o ~ ~ ~ b ~ w i t h w c o r d a ~ v i n g hfactfromBlOtlIfGtorganhWwhichhadada- ~ ~ b ~ c 0 d a e i m * ~ *

---P*-m f3irb mmmmn in both e ChiVal d m m u m p % mrfee. orgamhtioaai ~ ~ ~ u f a ~ w e l l ~ e d m r i m d f1- as annual reporte, minutes, exemtive dh&r'Ei c o ~ n d ~ , finartciaI r e c d a and al- Wdbafiangedprogramfi l -ah d d m . The ~epositorg crtn ph-y 6- the d m on the shelvee in an oder moving from the general to the specific, rsuch as from annual regorb to program flat, The tiom om d* u m to o r d e r ~ m ~ i n m o r e ~ i n t h e n e x t ebapter. mSe of the fil- will m f l d their p W - d wder on khe shelw. This very W c , but not uncommon, type of arrangement pmwmai both the provenance of the film as a whde and the htegrity ofthe filing 8tmdure thmugh the Beria @€m F'&m &I.)

TBe major WfImlth with tke syiWm derive fbm its me&m@ds on phpbl grouping and fixed hierambid &m&wee, Maiahmme of p w - cal unity and o r i g i d order im appropriate for the a c t u d f I b i + a ~ s ~ I t L n o t n ~ i n dam mts d records sharing a common prove 6. Ammgmemt by provenauce h indepdent ~ t h e ~ p r o x i m i t y o f f ~ c o r d s . A s t h e D u t & be& in 1- "by W p t i o n of an archival mIIw tionb a w e inventory, that oolldon from a ad- eptific mt of view becomes whole again. From that mt of view, it m a t h little where t b archival

mum 6 1 Roa& S h c h m M l d m d In Physical Fik -, Mmaam E d d o n Dtvislon

Repository Lewl

I I

Rtxord Groups

I Adminisidon €dueation D i m ... b g i h r

subgroups

O i m Childmn's Ad& Educuhl Div. D i i Program proOmms. . G e m m I k d s

1 4 ) (b -1 (b 7--ldj (b 44-45) w t

Annou- .. wbih " Youth Group Mumum Cwncil " Box 2, Folden 2440

Sub-subseries

h w l Rmprts 1 .

C o u d Minutw (Fddsrs 2733) . . . h M i

File Units (folders)

Item/Oocumd

January 1972 Minub, 3pp. "

60 Arrangins and Describing Archives and Manuscripts - -- - - - -- -- -- - - --- -

documents a re kept."The major practical reason for keeping records with a common provenance t qe the l - would be for user convenience in bl-owsing and retrieval, as in library stacks. Since archives and manuscript repositories have closed stacks, physical location becomes much less crucial.

The difference between the intel~ectualladrnin- istrative arrangement of records by provenance (in terms o f record groups, collections and subgroups) and their physical grouping and internal sorting by filing structure (in terms of series, subseries and files) has important irnplicatjons. In many ways this bwic duality is a more useful way to conceptuaIiz~ arrangement than the classic Holmes model. Ar- rangement by provenance on the record group/col- l~rrtinn and aubwoup levels has no necessary physical manifestation. This type of asrangemen t can be entirely on paper. Given the complexity and size of modern public and institutional records, such arrangement increasingly is only on paper, as an organizing structure for a large body of records shelved throughout a repasibory's stack area. Thus the 1967- 1980 park planning files should be kept together, since they arrived together, but they do not, have t o bc kept anywhere near n t h ~ r filps series from the Office of Park Planning, let alone all the other DNR records.

As t h e planning files illustrat~, where physical integrity of a coherent body of records exists a t ncces- sioning, it should be maintained to facilitnte processing and use. That will bc possible with many r~rords from old organizations or personal papefi accessioned as a group, such as the Southside records and the Bailey papers. IIaving arrived as complete bodies of records, the textual materials at least can be maintained together physically as well as intellectually. However, this is a convenient but by no means abso- lutely necessary situation. .4rchivish have jncreas- ingly recognized that their traditional idea of a unified hierarchy of levels nowitlg naturally from thc overal I repository organization down to t he individual documents was overbly simplistic.

Archivists now understand that arrangement by provenance and arrangement by Filing structure are two different hut related systems. Arrangement by provenance provides information about; records creators; arrangement by filing structure provides information about records. On the one hand there is the history of the Southside Cornunity Services and the biography of George Bailey; on the other the type of records which that history and that life generated.

' Muller. Feith nnd Fruin, .Wn~lzral. 12.

Figure 6-2 The Two Types af Arrangement

Arrangement by pravcnance

A r c l i ~ v o l rccard groups and mn7uscrlnt

C O C J ~ C I I I O R I

Subgroups ( n r c h ~ v e s and m a n v s c r ~ p t s ) , sub-

5wbyr~ups . . S a r ~ e s docurnent~ng act~v~ty / lunct ion

Arrangement by filing structure

Scrles 01 records flled t o g e t h e -

S u b ~ r ~ e s , stb-subser~es . F ~ l e units

lndrvtduor documents 2 n d ~ t e m s

Note: The some series w,ll or fpn rep7er;ent

orrangernen! b y both provenance ond filtrg strLc/ure

The difference is ns profound as the di f fe r~nce hp- tween information about. an author and information about a book by that author. The distinction helps elucidate otller cornplicntions within each of the two systems. Archivists are aware that complex institutional networks and relationships increasingly represent reality better than the simple monohicmrchi- cal model in which every office has one superior. Records in complex institutions may have no single creator or may have different creators over timc. The arrangement of modern archival records must be adapted to this reality. (See Figure 6-2.)

At the same time, as David Rearman and other theorists have noted, sets of records as such are rarely in a hierarchical relation to each other:" While a group of' records will have some type of internal organizing structure--most common1 y alphabetical or chronological-there will usually be no com- parable external organizing system dictating t h e relationship of those records to other records. While a program file marked "A" clearly belongs before " 3 in an alphabetical system, it is not so clear where all the files under "Programs" as a whole belong in relation te "Financial Reports," for example. Unless there is some built-in derivation or summarimtion of information--such as a series of indexes to a series of cases-records are not superior or subordinate to other records; program files are not at some inher- ently "lower" structural level than minutes.

T ~ P hierarchical model has even less relevance to the relationship between records and the offices

.' For a n au t s tnnd in~ discussron of the corlFuslon over hrer- nrchy In archival wr~tlng. we Bearman nnd Lyt le. "l'hr Power of the Principle of Frovenanc~." 1621.

Arrangement: TRvela of Control 81

that generated them. Minutes and other mr& do not "report" to their creating body in the way that body reporta to ita superior unit. The Iatter relationship can be illustrated on an organization chart, but not the f m e r . Records have relationships to their creating body and organizational environment, as well as to other recod, but these reIationshipa are not hierarchical.

Archival arrangement should thus not be thought of aa one unifid sy&m in which p h @ d files and file wriw are at eome lower level than m r d grouper, coIIectiom, and gubgmups. These are W d two different kinds of arrangemenkr- rangemerit by provenancelsecorda creator and arrangement by filing structure. Each works b& when

I separatd h m the other. There is no n d for a P-ed phydcal repduct ion on the shelm of a hierarchy extending all the way through an entire organizational structure and then file structure. With both types of arrangement regarded as more flexible, information about r e r d s and information about m- or& creators can be repmitented wparately and then related through links that ilIuminate the full range of their historical development. The complex history of the DNR and the mponsibility for park planning in Saratqp oomtitutea one kind of information; the descriptions of the various recorde generated by planning acthitie8 are another type of information. The relationskips among these different pi- of infofmstion constitute the framework for arranging and d b b i n g records. In an imperfect but ulseful analogy to librariea, we cern note that bibliographic information about a bwk brings together infoma- tion a h u t both the author, who may have writtm sevefal books, and information abut that specific

I book in order to make a complete dearription. In practice this perspective may have little h-

pact on the arrangement of many traditional records and papers. For praonal papers and many small organizational collections, especially coIlections no longer growing, the simplicity of structure often dlom a shultmeous phyical and inteIlectud arrangement. The arrangement of the B d e y papera would not n m r i l y require a gophisticated analy- sis of relationships b h e e n film and creators, nor wodd the few minutes and reporte from the old St. Joseph's Settlement taken over by the Southside Seb tlement. But even in thew cases it is useful to concep tualize physical and intellectual arrangement -pa- rakly. For more modern organizational reco&, it ia certainly more appropriate than the original system on which it is baed.

Levels of Control h this more complex environment, the aeriea

remains the mhst w f u I unit of anaIysis. AH noted in h p b r 1, a senha is:

a M y of file units or documents arranged in accordance with a unified filing system or maintained as a unit by the organization or individual that cream them b+ cause of some other relationship arising out of their creation, function, receipt, p h p i d form, or use. Arrangement is based on working with coher-

ent sets of record^ and relating t h w records to their origins and organizational context. &ria form the link between arrangement by provenance and arrangement by filing; structure b w they cam- manly combine physical inkgritsf with the documentation of an ongoing function or activity. Thus in the records of the Museum's Education Division, monthly reports of the various department heads not only constitute a coherent set of secorde, but also manifest a a p x i fic mpnsibil ity and activity. Series are thus at the center of the contemporary organization and management of records by (1) repositmy- level arrangement, (2) provenance, (3) filing structure, and (4) physical fzle unita and documents.

1. Repository-Level Arrangement. Many re- witaries have phy~ically and administratively - arate divisions to which different types of record p u p s or collections are mimed, Such arrange ment at the repxibry level i g concerned primarily with facilitating the logical and efficient administration of the institution. The mmt common divhion ie the divkion between manuscripts and archives in institutions which collect both. The Univemity W- brary may include the University Archlvm as well as the Regional History Center, rn that the first act of arrangement is to mign the Southside records to the Regional Hishry Center. A Bimilar division ie the one often made within institutional archives be tween organizational records and the pemnal papers of prominent officials or participants when those individuals donate their records to the ar- chivee.

Other divisions vary according to the nature of the repository, ita holdings, and ita clientele. Divi- sions on the bask of physical form (such as map, films and photographs) are common. In these cases, the r eps ib ry must keep careful separation records so that the intelLectud unity of colledions is maintained. Some repositories have ~eparate sediona of m r d group or coIlections devokd to time p r i d e

62 hanging and Describing Archives and Wzpscripks

(pre- versus postrl$0(P) or even topics, such as milf- tary histmy. All such divisions should retain the integyity of record groups, rather than divide them among different pa& of the archives. h manuscript repositories, vduable individual documents may go to a separate secure area. In general, the number of separate physical divisions within a repository should be minimized, b avoid wasting space.

2. Arrangement by Provenance. Though repository-level arrangement is convenient for administrative and physical purpc~ses, the most imporbnt organization of repository holdings is in accordance with the principle of provenance. Repositories organize their holdings according to origin and all records are linked to at least one creator. Where records arrive from that creator over time, arrangement by provenance is largely an intellectual and administrative activity. However, archives and rnanuacript repod~itoriee apply provenance in different ways ac- wrding to the sespective concepts of record group and manuscript collection.

As developed by the National Archives, a m r d graup is "a M y of organizationally related records esCabEishd on the bask of provenance with padicu- lar regard for the complexity and volume of the r e ords and the administrative history of the record creating institution or oqg~nkt ion ,"~ Everything after the word "provenance" reprmenta the practical modification of pure thmry. Yet the heart of the record group remains the old European fond, defined by Canadian aschivists as '"the whole of the documents of any nature that every adminstrative M y , every physical or corporate entity, automatically and o p n i d l y accurnulam by reaeon of its function or of its activity."" h a r d groups should be the fonds of administrative bodies with their awn rt+ sponsibiliti~, autonomy, and stability.

Most record groups meet this definition, which in our examples could easily apply to the Departr ment of Natural Resour- and the museum" JUuca- tion Division. In addition to such "naturd"goups, the National Archives created '"general" record groups for executiv+level departmental records and "'co1lective'"oups for the records of small but se- lated agencies. A general group could include the records of the successive secrehries of a cabinet department, but not the records of the functional units of the department, while a collective group could

Evans et al., "Basic Glossary," 428. Bureau ofCanadian Archivista, T o w a r d e k r i p t i w Stan-

dards: Reprt and Recornnwndntione I$ the Conarlian Working Group on Dencriptiue SludardF (Qttawa: Bureau of Canadian Archivists, 19851, 7.

include the records of many different but closely re- labd study cornmiasions d d e y d to the same general subject.

These adaptations are responees to the need to organk and control huge bodies of government archives. They emphasize the lack of rigor in the American record group concept. The American concept represents an explicit compromise between the logical implications of provenance and the practical realities of the varying size and complexity of records produced by different agenciea over different time spans. The defmition of a Record Group also takes into account the extent to which an organizational unit or agency has a separate history, autonomy and identity. In our example, if the Bureau of Parks and Recreation had generaM a vast amount of records and had enjoyed substantial autmomy, it might have been established as a separate record group. If all of the other bureaus were separate reoord p u p , the DNR mxretary's executive and administrative f d a might have become the equivdent of a gene& group.

The record group concept haa not been without influential critics including Peter Scott in the 1960sl and Max Evans in the 1980s." The justification of the record group an the basis of convenient size flies in the face of reality-- National Archivea rewrd group average several thousand cubic feet. More crucially, continual reorganizations of administrative stme- tures cause great difficulties for mhivw trying to maintain their records wing definite record group Returning again to our relatively aimple example, we may ask if the Park Planning files predating 1972 hlong to the Consenration Department record ~ u p - - s i n c e they were generated in that deparb ment-or to the new DNR record group? A plausible argument can be made for either. General archival practice assign8 the records to the DNR record group, since that is the most recent department, and the one from which the m r d s were t r d e m e d to the archives. ConfmntRd with such situations, some archivista have urged the abandonment of the record group sysbem in favor of concentrating on arrangement by series, with any larger adminietrative units described only as gmupingp of related series.

a The key asticlee are Peter Scott, 'The h r d Group Con- cept: A Case for Abandonment," American Archiviet 29 (Odobr I%), 4 9 W ; and Man Evans, "Authority Control: An Alterna- tive to the Rword Group Concept. A&n Archiviet 49 (Sum- mer 19861, 245261. See a h Mario Fenyo, "The Wrd Group Cancept: A Critique," American Archivwt 29 (April 1943, 229- 239; and Carl Vincent, 'The k r d Group: A Concept In Evolu- t~on" Archiuoria 3 (197677): $16.

Arrangement: hve la of Control 63

Adaptation and mdification of the record p u p mricept is generally preferable to complete a h d m m e n t . For archival administrative pur- powa, recoda should be identified, even if arbi- tranly, with gome creating or maintaining agency. Changes, mergers, and dl kinds of adminhtrative relationships can be represented through a variety of descriptive elements such as hietorical agency name Ihta and organizational charts, as well as traditional index- and references within narrative descrip tions.

fn such a context, the atrength of the record group m m ig apparent. For most records, it d m accurately represent on the large acde the organization which creak3 them, whether our M w u m M u - cation Division or, a t the National Archives, an agency like the Federal Aviation Adminigtration. The record group also remains a more useful unit for overall archival management, organization, and d b p t i o n than thousands of separate aeries. Many of the latter will have a general and relatively unin- formative title Iike "Case Files" in the absence of the traditiond record p u p information that would commonly identify their provenance.

The orgamktion of the hoIdings of manuscript repositories prdominantly in terns of "organic" col- Isctions differa from the clamic archivd system in ways not always fully apprmiatd. Collections of organizational m r d s and pemnal papera established on the bash of provenance are wmmonly conceived aa the amlogues of record groups. But in fact s u ~ h mII&ions are of'ten the entirety of the records of their creatum. Thus, ail the recorda of Southside Gommunity Services (one manuscript collection) are, in fact, equivalent to all the records of the Museum, not juet the Education Division record p u p . Such mauuacript oollectiona thus represent a direct and

4 uncompromisd appIication of the principle of prove- mnm. In addition, since many manuscript collections came intact BB a whole from homes or offices, p n w e m c e can in thme cases be maintained physically as well as intellectually.

Both manuscript collections and archival m- o d grouw can be divided into subgroups. Subgroups rep-nt admhhtratively discrete units or activities that produce records. The basic difference between a m r d pup/collection and a subgroup ie that all subgroup have a common origin in record mupa or collections, while record groups and collec- tiom should represent reasonably autonomoue agen- eiea or organhtiona. Howwer, NI nokd above, an orgariktiond subgroup that is 1- and indepen-

I

dent enough can be establishd as a separate record group, according to r ep i to ry policy.

Subgroups are sometimes mnfueed with =rim, h u ~ e they are often small enough to be documented in one set of files. I t is not at d l uncommon for a small office-an archival s u b g ~ o u p t a have generam only one seriea, such as minutee, which have enough vaiue ta be retained in an institutional archivm, Subgroup denote the context of m r d s creation; s e r i ~ are the recbrds themeelvw. &en if the s u b ~ u p and the aeries are physicaIly identical, as in the cam of the minutea, the distinction i~ important conceptually. In fact, a single minute b m k representing all the sunriving recorda of a d e w ment can be simul~neously a subgroup, a wriee, and a file unit.

In large organizations, there cam be many levels of eubgroupa, sub-suagrOups, and ao on reflecting the administrative structure. S u b p u p can also be establishd for the records of predmessor, merged or relatd organizations ahorbed during the organiea- tion's history. Activities and functions ahould be grouped where a number of related series would have no other gtructural Iink. If an individud'e papers Iack any formal structure, archivists sometimes create subgroups of related seriea of busin=, civic, or pemnal activities or functions.

Subgroups, like record groups, present the same problems of constant flux in active organh- tions. Such change ia even more common at the office and program level of erubgrouping than at the higher organizational levels. As a practical matter, sub moup should therefore be limit& as much aa possible to fairly stable units and function%. A business archim, for e m p I e , may create subgroups on the basis of continuing functions such as production, marketing, personnel, and fmancea.

Our sample collections illustrate gome of these considerations. The simpled case i~ the mwum'e Education Divieion. The recorh of organizational. units within the division auch BS the International Prcgmrn and Children's Program8 form obvious sub group. In the w e of the Public Relatiom Depart- ment, loc~ted within the Education Divieion until 1965, crase-referencm in different dmxiptive tools will link thb recorde subgroup with the poet-1965 Public Refatione Divieion record group. But the pre- 1965 records should remain part of the Education Didion'e historical materialn. The Education Divi- sion itself mporta to the aesociate director for administration, whose tecards form a epamb record gr0UP.

64 Arranging and Describing Archives and Manuscripts

The 1967-1980 park planning files are more complex. For administrative purposes the files would be part of the DNR record group, with cross-referenc- ing to the pre-1972 Conservation Department record group. The files wadd also be identified with the Bureau of Parks and Recreation subgroup. The ques- tion would be whether to establish the Office of Park PIanning as a separate sub-subgroup. If the office's only records were one relatively small series sf planning files, they might be treated as a series within the bureau, though they could be treated as a subgroup which happened to consist of just one series. Since in this case the Office of Park Planning has been in existence since 1958 and has several other series in addition to the basic park plans, the office records become a subgroup of the bureau records subgroup. This sub-subgroup will not, however, include the records of the park planner on staff from 2935 to 1958 before the office was created, since these are already organized as a series within the records of the old Conservation Department. (See Figure 63.)

The manuscript co1lwtions present different challenges. One set of recor&+those of the Whomso- ever Mission-found at Southside Community Ser- vices was identified as a separate collection during accessioning, because it always functioned ag a separate organization. To the extent they can be identified separately, records of the various predecessors of the current agency will be established as subgroups, including the records of the two settlements a b sorbed by Southside Settlement, the Day Nurseries agency with which it merged, and the direct precur- sors of the agency such as the Union Mission, South- side Settlement, and Southside Community Center. In addition there will be subgroups based on the agency's recent structure, such as the Youth Ser- vices Division and the Adult Literacy Propam. These subgroups will serve as a conceptual framework, though associating specxc series of records with the various subgroups will be a complex task.

With the Bailey papers, the major issue will be whether any subgroups are needed. Assuming the records are in a sensible filing order, a further level of structure may not be required. Subgrouping might be convenient if there are many small sets of folders each documenting a different civic activity or financial matter. A more interesting issue would be the disposition of Mary Bailey's correspondence and her files relating to her work with the museum, If easily separable they could constitute a separate collection on the basis of a separate creator. However, if they are integrated into a much larger corpus of George Bailey's files, such a separation would violate prove-

nance in the sense of the records' origins and maintenance, as well as violating original order. Depending on the contents of the entire collection, and the opinions of the donors, the collection could ke identified as George Bailey's Collection of Bailey Family Pa- pers. This type of decision h o m e s the province of policy and judgment.

3. Arrangement by Filing Structure. Archi- vists organize and manage physical groups of records primarily in terms of series and subseries which they identify or create. Arrangement by provenance has no &re& relation to the internal order of individual series or the arrangement of series with respect to each other. Arrangement by filing structure, in contrast to arrangement by provenance, is concerned with the reality of sorting, grouping, and shelving the records themselves. It is also concerned far more with the starage, handling, and retrieval require ments of records in different physicat formats. Since some series anive zts accretions over time to ke stored in various lmations, arrangement by filing structure is not always a physical arrangement. However, the physical aspect is usually much more important than in arrangement by provenance, The language of provenance is a language of creators-- Office of Park Planning, George Bailey, Southside Settlementwhile the language of series is the actual language of r ecorddar i e s , minutes, ledgers, correspondence, or grant proposals,

The standard defmition of a series noted pre viously describes two kinds of series. The first is based on a coherent filing system in which a set of records is arranged alphakticdly, chronologically, geographically, or according to some other consistr ent classification system. Such series include the common and easily recognizable record types like minutes, reports, correspondence, Ease files, and accounts. (See Figure 6-4.) The second kind of series is a set of records which may be a mixture of fiIe types and arrangements, but which derives some unity from either a common format, such as the park planner's maps, or from a common function or activity. These may be records relating to a subject, a p r e gram, or a branch office, and may thus resemble subgroups in that they will consist of different types of files. A series entitled "Planning Review, 1980" might bring together several kinds of records, but their creation and use in a specific activity will make them part of one series. In disordered collections, archivists can establish all of these different series as needed.

The series is the key unit for pr-ing because it combines both the file structure of records and the

Ffgurs 6-3 Records Creating: Park Planning Inrtitufional Canisxtr, 1967 mnd 1980

1967 Department of Consewdon

Deportment of Conservation

I I

I' Parks and Recredion Commission Fish and Game Commission

I Forks Historic Manummts Lokes and

Division Sites Bwreau Bureau Streams Division

1

I

I

Maintenance

1980 Department of Natuml Resources

Department of Natural Resources -I**

l nfergovernrnenlol Planning Group

Office

*** %tote Ermironmentol Qualify Council

Division CulturnF Sites Recreation

I Management Securify Services Special

Programs

D i d Reporting

-----I- Coordinating - Provision of Planning/Suney Services

I T I Assistant Secretory for Assistant Secretary for Assistant Secretary for Outdoor Recreation Resource Management Administration

I I I I Bureau of Porks Bureau Forests and Mines and Land and Recreation of WildlZe Weilands Miner$m-4-.-d Management

I Management I

I - e m C C

e- l lL-- - -L m

1 1 I Park Planning Porks Historic and Woter

68 Arranging and Describing Archivw and Manuscripts

FEgura 6 6 Camman f ypss of Functional Remrds Series

Memo-ando

Motion Pictures Newsletters

Notebooks

Aufhorizotions Otganizotional Chotts

Bank Staternsnts

Press Releoses

Prodornotions

Correspondence Publimtions

Recornmendotions

Research Notes/F~Ies Resolutrons

Schedules Financial Stataments

Histories Speeches Indentures Storernents

Indexes Statutes Invoices Studies Journals Surveys 'lows Testimonials 'lectures 'restilnony ledgers Tronscripts Legal Opinions Videotapes Lntters (sentJrecsived) Vouchers Letterbooks Warw R ~ S

Logs Wills Monuols

documentation of a record-creating activity. Experi- ence telIs us that activity ia often better described in terma of tangible serieeminutes, tax lists, memc- retnda-than in terms of constantly changing bu- reaucratic facades. A series represents the combina- tion of physical file integrity and a spec~fic function or activity. In this way, the integrity of a series pre- sen= the values of both provenance and original: order. Each type of record at the series level has its own characteristics, which archivists should understand. By their nature, minutes contain a dserent kind of information from annual repom or staff memoranda; annual budgets are different from ledgers. This recognition of the importance of dimrent types of records as conveyors of different kinds of information further enhances the central role of s rim in archival processing and use.

Archivists denote subsenes where there are si nificant and logical divisions within a series. S u b rim must be an integral physical part of a =rim maintained as a unit. There are two major typs of subseries. The first occurs when a aeries is divided into two or more parts, as in a correspondence file divided into "Lettera Received" and "Letters Sent,'" or a unified research file divided according to different topics. The second occurs, when subsets of files are "n~ested" in a larger set. Am individual's "Finan- cial Records" may be arranged alphabetically, but a t the letter "T" a whole chronologidly arranged set of tax returns may appear. These tax returns constitute a subseries. Subseries almmt always derive their existen@ from the way they are physically embedded in the larger series structure, Original order in such situations mems physical file order. (See Figure 65.)

Unlike the relation of subseries to series, the ordering of wries in relation ta each other is purely arbitrary. This issue relates both to the placement of records physicatly on the shelves, and to the listing of records in archival inventories and other finding aids. Traditional registry sygtems usually provided a fmed structure for all the different series p d u c e d by an office, and the retention of that structure was an integral part of the retention of "original order.'" Such registry systems have been rare in the United States, however, and archives commonly acquire Be- rim that have no predetermined relationship to other series from the same organization.

On the basis of the principle of original order, when records arrive from. an organization or individual in a wt order that is not purely capricious, that sequence should be retained. But there is no inherent "'organic" link or ranking that d i c t a h the physical

Arrangement: Levels of Control 67

Figure 65 Ssrin ond Subseriss, Purk Pknnlng Aisr mr Trunstsrred from Agency -

W m t 8W67--Park. Planning F i k { s e r i ~ ) b

Box ff Sub8erles Sububserlmr Suhububssdm8

1-4 Departmentol/Divisionol Annuol Planning Summaries, Files 1976-80 [Box I , Folders 1-5)

. . . . . (Other Files)

S 9 Stota Park Plans Apple Volley S. P. Renovation Budgets, 1 9 7 6 7 9 (Box 5, 1 Folder 1 ) . . . . . (Other

(Other Parks) Aspects)

10-13 County Park Plons Adorns County Harrisville Nature Preserve, 1 9 3 S 7 7 (Box 10, Folder

(Other Countiss) I ) . . . . (Other Parks)

14-15 Pork Mops Stote Parks (Box 14)

I Coun!y Parks (Box 15) I order in which the series constituting a collection should be arranged. There is nothing to mandate whether or not staff memoranda blong before or &r financial records. Attempfx by archivists b de- vise principles for such relationships were an out- growth of the system which joined arrangement by provenance and arrangement by filing structure in one u n i f d hierarchy, The ordering of seriw was the link between the two different types of arrangement, eince it creahd the impmion of a hierarchy of filing Btructum immediately subordinate tn the hierarchy of mfd.creatinp Mies. But as we ahall see in the next chapter, the arrangement of =rim is a matter of mnvenience and judgment, not fundamental archival principles. It is essentially an estimate by the archivist of the relative scope and impor&nce of the various aeriea.

?n our model t..ecorda, the park planning Filw of the Office of Park Planning reprent a fairly typical series. The 1967-1980 files are actually an addition tb the already ezisting wries of 1935-1971 park planning filea &red elsewhere in the archives. The four- year overlap btween the two aeta of files ie typical of situations in which individual fdes remain active for wrying lengths of time before being traderred to atorage. The 1965-1980 files will form an addition to the e M i g series, rather than a new series, be- came in our scenario they are part of an ongoing fde structure, with regular chronological breaks. Within the new addition there are records which will be n t o d separately from the standard paper files, such m tap, photographs, and audio md computer t a p . As hae been h r d , thie does not reflect on the intellbua9 integrity of the series. Their common

provenance will keep the bxtd and nontextual records linked administrativeIy and intellectually, regardless of separate storage or even separate index- ing and retrieval systems. The =rim also contains Files which represent activities carr id on and documented by other qpnciea before the government re- organization of 1972, but these files remain with this aeries. In the Office of Park Planning, the series of planning fdm was fonnally maintained in four sub divisions-DNa/Division directives, atate parks, county parks, and map; these are maintained as mbseries,

The mwum's Mucation Division records flw trate a m r d gruup which was tmferred as a unit, so that the s u m u p s , eeries, and subr ie s are in the physical relationship to each other that was established by the division. Even if there are unpre- dictable sequences of file p u p (perhapa Interna- tional w s has been placed before Children's Progmnm, or grant proposaLs ahead of minutes), it is best to retain this eubgroup and eeriedevel order. Here the phpical and intellectual arrangement can be one.

The most diff~cult challenge will be the identification of seriea within the Southside Community Senrim rmmrda. Changes in organization and mission, in record formats and types, and in directors and administrative secretaries over a century could d y produce more than Flfty identifiable aeriee in fifty boxes. The Day Nursery, for example, mjght have retained only t h m cubic feet of records, but these could include four series--minutes, program files, scrapbks, and files of the Middletown Day Nuraery Aesoeiation. Some of the other mriea could

68 Arranging and Describing Archives and Manuscripts

extend over the course of m r a l organizational changes, complicating the relationship of %ria to subgroups. A consistent series of reports to the United Way since 1940 will carry successively the names of Southside Settlement, Southside Center and then Southside Community &Services. With small sets of folders the archivist will have to decide whether to view them as sukr iea of larger series w group them into programmatic subgroups, such as 'cHealalth Projects." For the totally disordered records dating 1968-74, the staff will have to create series based on a atudy of b t h the pre1968 files and the we1 l-ordered posb1975 records.

G-eorge Bailey's files more closely resemble those of the Museum Education Division. His personal papers as transferred from file cabinets were already in file groups which can fairly w i l y k corn- bined Iby the repository into subwups ("Civic Af- fairs"), series ("Personal Correspondence"), and even sukries (Mortg~es by year withn "Finances-Per- sonal"). While there will typically be mme folders or documents difficult to categorize, the coIlection as a whole will retain its full order inbct.

4. Arrangement by Physical FYe Unita. Olzginal order is applied most directly in handling physical file units such as folders or bound volumes. In contrwt to serle, whose constituent parts can be in different parts of a repository, file units really can be in only one place at a time. The order of the units im- by the creator is the very structure which defmes the film as a series. Generally that order is chronological when the series documents a regular activity, such as minutes of meetings. The order may ke alphabetical when it docurnenh programs, bpi=, organizations, or people, such as case files arranged by the name of the client. The arrangement can also be by some internal classification ~ y s t e m , like a deci- mal subject claasfication, usable if the archives can find a codebook or clagsification scheme in the m- ods. Series created by an archivist from loose papers will usudly be arranged chronologicalPy or alphabetically, according to the type of function or activity they document.

Within file units such arr folders, individual dmuments will also be in some chronologid or alphabetical order which should be retained. In practice, the holdinga of archives and most manuscipt r e p i - toriea are so Iarge that archivists rarely review item-

level arrangement. In addition, in the absence of close mpxvision, use itself often causes documents to b m e disordered within folders. Only where there are individually valuable documents should the a-c location of such item be a major con- cern. Detded arrangement usually ends with the ordering of file units, not their conknh. The major exception ie the case of records which are to be m i c r ~ filmed. Such m r d s must be in a precise and l o g i d item-level order, since images cannot be reshuffled or viewed more than one at a time.

Our sample archival records raise few problems of file-unit arrangement. Typically the only park planning textual records needing detailed arrange rnent are the repom on apecific parks from the park planner's office. These will be used heavily and ueem will ask for individual items. In contrast, the South- side recards arrive with boxes of unsorted lmse papers from 1968 b 1974. They also include many bound volumes which need b be brought together by provenance and type (Settlement House minutes, camp ledgers, etc.); they then must be asrangd chronologically. Several series in the Bailey papers might be aman@ in some detail. The autograph series will certainly have to be arranged alphabetically and listed by ikm. This is the procedure the historical society would use with individual "artifi- cial callectiom" donated by collwtors. In addition, the general comspndence geries might be arranged alphabetically by incoming correspondent since that would be the way documents would normally be re queated, bs with the other recoda, the detailed arrangement of aome series does not imply the detailed arrangement of all wries.

The different types of arrangement desmibed in this chapter provide the framework for the organization of records from divisions at the repository level through the placement of individual d m - ments. That framework allows archivisb to mange records in the distinct ways appropriate to the preservation of provenance, filing system, and physical integrity. An d k d in the next chapter, it also provides as well for the organization of completely disordered records. The arrangement of any specific body of records is a function of ita own physical and intell&ual characterktica. But the practices and p d u r e s of arrangement take place within the larger system reviewed here.

/ / Chapter 7 / ?

I

I 1 Arrangemen): Practices and Procedures

Because both records and repibries are sa varied, arrangement in practice can be a whole range of activities from simply approving a received filring system to establishing order in a mass of disordered documents. Whatever the initial state of a set of accessioned records, the procws of arrangement ~hould accomplish several gods. Records should be physically grouped inb coherent series and subseries. Within those file structures individual file units will bz either in the order received or in an order &blished by the archivist. Records should be identified with their administrative and/or functional origins, as part of larger record groups, manuscript collections and subgroups. Whether thme identifiea- tions are purely on paper or a h manifestd in a physical arrangement will be determined largely by the nature of the records, Manuscript collections are more likely than complek archival record groups ta be maintained as physical as well as intellectual units.

Arrangement in public and institutional archives is often a p s m of preservation of original order. In argmhtions with well&blished records management programs, most r m & arrive in a wmprehensible order from established functional entities, Archivists should verify that order and en- mre that they have r?o& information about the creator or creatom of the records. The creator will in many caw^ differ from the immediate source. In the majority of cases where these issues raise no problems, the basic goals of arrangement have been met. Further work (reboxing and refoldering, and listing file headings1 is optional and often clerical.

The arrangement of manuscript collections very often involves much more detailed work at the folder level and measionally even at the document level. Here archivists m a y have to create, replace or reconstitute a filing structure. Even well-ordered manuscript colledians commonly contain some disordered or misidentified records. Because 811th cal- lections offen represent the entirety of an organization's or individual" records, rather than particular series, the staff must deal with b t h arrangement by provenance and by filing structure, Manuscript curators as well as institutional archivists can also be confronted simultaneously with changes in record-creating activity, organizational structure, and the nature and content of records, further complicatr ing their work.

Manuscript curators may be less likely to re bin arbitrary or idiosyncratic original filing order, since manuscript repositories concentrate on satisfy- ing external research n& rather than the needs of intesnd administrators. The most common challenge in manuscript arrangement is to create from disoralered c~Elections a complete and detailed arrangement based upon provenance-an arrangement which reveals the reasons the files were created. In addition, individual documents of special, value are more likely to be found in manuscript re- posibries than in aggregations of institutional and government records.

Despite these differences, there are impartant similarities between archival and manuscript arrangement. Archival records frequently contain some disordered files, while manuscript coElections

70 Arran#$np: and Describing Archives and Manuscripts

received from contemporary organizations resemble institutional archives. The overall goab and sequence of work of archival and manuscript arrange ment are aimilar. The ultimate goal is an arrange ment of records that will permit their effective use by facilitating description. The mume of work at all repositories can include the following stages: preparatory work, arrangement by provenance, wries- level arrangement, rearrangement, arrangement of fiIe unita and documents, and physical handling and storage.

Preparatory Work Before arranging a group of records, an archi-

vist should h m e familiar with their origin, contents, arrangement and condition. Much of this information should be available from the various form and summary description8 created during the a-ition and acemioning stages, However, the arch id should generally do some further research into the hishy and context of the creators of the records, and the activities documented. Such re search should not be exhaustive at this stage, but could involve reviewing internal histories, organization charts, biographical dictionaries, local encycl* pedim, and other standard reference works. The archivist should aim become familiar with related records held by the repsiaq.

Research sets the stage for the initial examination of the records and any accompanying descrip tions. A review of the boxeg themselves and what ever Iiats have been created provides a gobd summary of the contents, arxangement, and physical condition of the recorda as a whole. The archivist should take notes on existing file structures, obvious gaps in the records, loose or disordered documents, and the types of records which usually require pr* cessing at the item level. B a d on previous research, the archivist should be able to recognize maor sub- jects and events, imporbmt individuals, and the overall strengths and weaknesses of the records in terms of content. Major organizational changea not mentioned in aecessioning recoda should also be noted. ks preparation for further work, aids to processing such as annual reports, internal histories, autobiographical sketches, and organizational charts can lx identified and even copied for use by the archivist,

Following the review of the records, the archivist can determine their specific p m i n g needa. Thh can take the form of an explicit work plan ( m x Chapter 5), The plan will require the archrvist to

explain and justify the proposd arrangement, the associated staff aasignmenta and mquird resources, and the impact; of the work on reference and conservation.

depend in^ on decisions about the nature of the m r d s and their processing needs, the archivist may conduct further sesearch before beginning arrangement. If the records as a whole are in disorder or are unusudly complex, the archivist may hare to function for a t h e as an institutional historian or biographer. Only with a full undemtanding of the context and activities that generat4 the records over time can they b properly arranged and described.

The nature and order of the records alm dic- ta- the quence in which the work of arrangement p-. When recar& arrive in order with a coherent structure, arrangement work proceeds from the largest unit to pmgremively more detailed subdivision~. In such situations, arrangement eonaista mainly of placing record containers on the ahelves as received-reboxing them into archivalquality containers where n w - a n d recording infar- mation abut each record-creating ndrninistrative Ievel and suMiviaion. This help the archivist to prepare for the descriptive work to follow.

In our m e u r n example, information about the Education Division record group as a whole wouId be derived by the archivist from an overview of the divisional records as summarized in a-ion forms. The archivist would then review the information about the records of each of the departmenb which form the subgroup. Finally, the individual series documenting functions and activities within the department subgroup would be identified. It would be v i b l e for the whole prccem to p r d without any physicd rearrangement of files. Arrangement of such ordered rwrd group can easily proceed to a given level of detail and then atop if necessary. The archivist may further refine the recorde later. (See Figure 7-1.)

Most collections of personal papem and dieor- dered or complex organizational records are not so amenable to arrangement from the top down, The archivist in these cases must identify the coherent file wts which form series and then d a t e those series with recordereating bodies. Once established, the ~eries will aleo provide the structure within which folders and dwumente can be placed. The Southide Community Services records typify digor- dered and complex collections. With no overall order ta the r w r b , the archivist cannot work from the top down. Imtaad, the varioua cries, vo lmw, and

Arrangement: Practices rand f rocdlures 71

Figure 7-1 Museum Education Divisicn Records, Unified Intellec)ual/PhysicaI Breakdown

other documents must be assembled, and the overall order produced as a result of arrangement. Here, the work of arrangement produces an order rather than an existing order prducing an archival arrangement.

The different approaches are based on the dual roles that series play-they represent file structures or t y p s as well as dmument activity, Both of these functions should be respected in those situations when arrangement must begin by establishing series rather than by identifying series within subg~oups. Thus in George Bailey's papers it would not IE appropriate ta create an all-encompassing correspondence series simply on the basis of record type. If Bailey maintained his correspondence in separate groups by activity, there should instead be a separate correspondence series for each of Bailey's activities. There might be one correspondence series for his civic affairs and another for his business affairs. Con- versely, if he grouped different types of records from different creators together in dmumenting some civic activity, that activity should be the basis of the series regardless of the mixture of sources and record

types. Thus his files on the 1956 Republican National Canvention might include correspondence, local cornrnittm minutes. travel and expense documents, a convention program, and a signed thank-you note from President Eisenhower.

Some manuscript collections are so disordered that they must be painstakingly reconstmcted from the documents up, reversing- the whole process mi- cd of archival groups and series. With collections of more than a few dozen documents this is naturally very time consuming. However, it really is unavoid- able if the materials are ta be available for use. As discussed below, the whole collection must first be examined in order to identify the different types of records and possible series. Once a framework of series is established, individual documents can k~ sorted accordingly.

Subgroups

Diviseon Director

Children's Progroms

I Adulf Programs Deporlrnent

School Districl Programs

Arrangement by Provenance

Boxes

1-1

5- 6

7-14

15-28

Ssrles 1 Boxer/Foldarr

Records should have been identified by record group or collection during accessioning. However, this identification should be confirmed, and such ba-

Volunteers Council

International Programs

Adm~nistrofive Secretory

Public Relations Deportment

Education Division

Administration Programs Personnel Correspondence Reports

Progroms Budgets

Memorondo Stoff

Publiciv

I

2-3/1-15 3/16-27

1 3/27-39 to 4 5/ 1 - 12 5/13-31 5/32-42 6/ 1-8 4 / 9 - 14 6/15-30

29/1-18 ....... 36/1-9 ....... 40/1-16 ....... 43/ 1-27 43/28-4 1

44/1-25 44/26-38 451 1-22

Reports ....... Annual Programs .......

29-35

36-39

40-42

43

44-45

7/1-9 ....... 1511-76/7 .......

Council Minutes ....... Reports

....... Div. Accounts ....... General Files Publicity Lists

Minutes Annuol Reports Publicity

72 Arranging and Describing A mhives and Manuscripts

sic information aa dates and volume verified. The remrds may alaa receive a record group or collection numkr at this stage. The record group or collection number d be different from the aotlessjon number. It will be wespecidly important if more than one ae- -ion is to te processed together as a p u p or wl- l h i o n . The record group/collection number will be used in all future processing and reference work.

Mwt of the work of arranwment by pmve nance concerns the identification or establishment of subposps. It is irnporbnt to recognize, however, that not all mrd group or oollection8 are neceasar- iIy divided into subgroups, nor do all series have to be clwtered into aubmups. Subgroups are used by archivists only when their existence is juetified by organizational structure or some tJnpe of general function or activity. Subpups derived from eKisting organizational divisions should be maintained separately from each other when the records are boxed and listed. The records relating to each subgroup will be listed along with the reason for the subgroup's creatian. With very large organizatiom, common wnse and r e p i t o r y policies should dictate when to stop creating subgroups for every admin-tive unit or type of organized activity. At some point, r ecod of activity can be established as series rather than as subgroups, even if the activity baa some organizational manifmtatioa. The of Park P l m ning will have a separate BtaEf member for county parks, but since the permanent records generated by that individual are relatively few and form part of a larger fde structure on planning, there would not be- a separate County Park Planning subgroup.

Subgroups created by archivists, especially in work with mm&pt collections, may lack the physical unity of mbgroup maintained by original m& creahrs. Certainly the aeries meatd by archivists will be in no preestabhhed order in relation to each other. Instead, the archivist will ereate sub p u p s on paper by listing together series which have some common relationship of origin, form or activity. But this work depends upn , and follows, the arrangement of the records thernselrea into coherent eeriea. (See Figure 7-2.)

Series-level Arrangement

The o r g h t i o n of files into series is the cnu. of all archival arrangement. It involves grouping rec- & into coherent units, placing them intol a provenance-based context, and ordering the series in relation to each other. There me eaaentidly four rnethoda by which series are generated-identifica-

tion, creation, recomtitution, and rearrangement. AI1 but the fm include mrne physical manipulation of records. The identification and maintenance of ~ e r i e a in a m b l e order is the easiest alternative. It becum in many public and institutional a r c h i v ~ where arrangement pr& through a p m of successive subdivision. The creation or mnstitu- tion of gerEes is more common in manuscript work. The tedious work of rearrangement ia needed when either archival records or rnanwripte- arrive in an mangement that is i m v i b l e to interpret or i~ pas itively counterprductive to research.

The identification of mriw in well-ordered collections ia generally a matter of h r n i n g the obvious coherent Filing p u p a , sometbeg with the ass& h c e of an internally produced file eystem outline. Breaks in Siles m u d eolely by clerical action are not jusW~rntions for the establishment of separate genes. Lf n w , such problems can b handled as subseries. Suhriea should also b u d when there are two interdependent sets of records, such as old aettlernent house regidem divided into eeparate h o l m for men and women. Archiviste should follow the practice of the office of origin in such determina- tions. The park planning files were always maintained within the office as one series with four BUM- visions, so the &ate archives will not tv to make separate series out of the county parka files and the state parks fdea, even though these record sets have the characteristics of series.

The creation or reconstitution of series in disordered colI&ions is EI much more mmplex task. The archivist f a d with such a collection has ta survey it completely, recordhg the t y p , dates, order, and condition of all groupings of Wes and all uncld- able film. In addition to ~ u c h details, thia survey will provide a centext in which to view the creation of possible subdivisions. It will often be nmemary to compare file headings with fde contents thoroughly. Relying on any lists prepared during a-ioning and without physically moving any recoda, the archi& can prepare a tentative arrangemen t of series and subaeries. The proposed arrangement will be reviewed with coIleagues and further refined during processing.

Series will Z>e creatd on the h l s of type and activity as previou~ly d k d . A chronological order will be recreated or established for series such as minutea or periodic reports, while program f d ~ and carreapondence may be aortal alphabtically. Other internal -1118, such as a numbering system for case files or a geographic d i n g for counties, will be reatabliehed where comprehenaibla. The trained

Arrangement Practices and Procedures 73 - - - -

Figure 7-2 Series and Subgroups--The Bmiley Papers

archivist" knowledge of the general characteristics of different types of series in different types of coIlec- tions can be of vital assistance in identifying potential series in disordered collections. Archivists should be familiar with the general principles of records management as well as with the historical evo- lution of rndern file creation and maintenance prac- t i c e ~ . ~ The identification of potential series in a

'Some standard records rnenaEement texts are Mins M. Johnson and Norman F. Kallau~, Records Mona,qerncnt, 4th ed. (Cincinnati: Southwestern Pu bllshing Ca., 1YK61; Violet Thomas et a!., Remd.9 Managerne~t. $s tem and Adrnlrrktrattnn (New York: John Wlley & Sons, 1983): and Patricia F. WniIace et sl., t?rnd< hfanafi.~rnenl-In l ~ g m ted I n formalton SLvxs l ~ r n s , 2nd 4. (New York John W i l ~ y & S n s , 19871. Among the Few articles on

1

I

disordered collection will be assisted by a knowledge and examination of similar types of records and record-creating bodies.

As in the creation of archival r m r d groups, some practical considerations enter into the creation of series. In the Southside records, if there were two

in fils cabinets hemdings saris. assigned

Middletown Republican Porty Political Activrtres

Sorotoga Republican Porty Political Activifies

Republican Notionol Conventions Political Astivi t ies

Middletown Chamber o l Commerce Civic Activities

Middletown Chamber, Executive Committee Civic Activilies

Bedford Falls 5ov1ngs ond loan Business Activities

First Notional Bonk Business Activities

Middletown Rofoty C~vic Activities

Middletown Elks Civic Activities

( 1 4 Middletown Groups-l file each) Civic Activ~ties

(8 Sorotogo Groups- l l files) Civic Activities

Businessmon's Club Civic Activities

Bankers' Roundtable Business Activities

Southside Community Center Cnic Activ~ties

Southside Community Center Board Civic Acfivit~es

Southside Community Center. Choirmon Civic Activ~ties

Middletown Museum of Art Mary Bailey

Diories-Peter Barley Peter Bailey collect~on

Diaries-Mory Boiley Mory Bailey

Correspondence-Family Personal

Correspondence Personol

Scrap books Personal

Soratogo State University Personal

Elm Siree? House Personal

Lake Winomonowoc Properby Personal

United Fund Civic Ac t iv i t~e~

finonciol Statements Personol

Bonking Reoding Files Business Activities

Speeches ond Lectures {To be divided)

Miscetloneous (To be divided)

the history of file systems are Joknne Y a k , "Internal Communi- cation Systems in American Business Structures: A Framework to Aid Appriaitial," Americnn Arrhiuist 48 (,Spring 19%) 141- 15% JoAnne Y n M , Control t h r o u ~ h Corn munlrations tRaltimore Johns hop kin^ Pr-, 1989): find JoAnne Yntes. "From P r m Book and Plpeon Hole to Vert~cal Filing: Revolution i n Storage and A- Tor Correspondence," JournaI ofRusiness G m n i u n ~ c a f ~ l n 19 (Summer 1982): 5-26, Chapter 9 of Schellenberg's Modem Ar- chrue,~ #Ian discuses filing.

74 Arranging and k t i b i n g Archives and Manuscripts

suwiving folders each dealing with the Settlement's nursing, dental, health ducation and medical clinic activities, they might b m e part of a series called "Health Activities"; but if there were a cubic foot of records on any one of those activities, it ww1d be constituted a ~eparate series. Such adaptations are especiatIy Imporhint with small mllwtions in which each folder or record type could theoretically constitute a wries, It b not always necessary to divide all the files of a small collection or subgroup into series. Collmtians of just a few h x m would often do better with a simple alphabetical and/or chronoloFfia1 arrangement. Personal papem consisting of one lmx including a few diaries, m r a l dozen lettern, a serap book, some memorabilia and various financial records such as a deed and a will would not have to h formally divided into a halfdozen wries. All that would be needed would b~ to organize and list the F f l ~ by type and then chrono~gicaIly by file unit.

The u~ual goal of arrangement Is a collection physicalIy divided inte diacrete series or subseries to illustrate function or activity, and assembled, at least on paper, into different subgroups ta document creating organizational unib. In the Southside collection, there may be a halfdozen groups of records which document the camp program over the mume of frfty years, including its seveml changm of organization and Iocation. Some record sets are obvious series, such as counselm's reports and lists of eamp ers, while others might be a few folders each on various programs or activities scattered over the years. All such groupings have to be sorted and identified

wries, subseries or, for all the camp records b gether, a subgroup.

The final stage of geries arrangement is h~ record some basic identifying information about the s e ria. In mll+rganizd proceesing pragrama, this activity will draw upon work done during accessioning, and help lay the foundation for the more elaborah descriptive work to follow. Basic information to be recorded abut each ~eries includm creator, title, time span, volume, and arrangement, as well as a summary description of its contents, Series titles supplied by either the creator or the archivist generally combine a function or activity with a type, as in Director's Compondence or County Plan P r o p als. Archivista often assign or supplement titles where the identification supplied by the eresrtar would be either misleading or inadequate for users.

In identifying creators and titles of records and papem and indicating their volume and date span, archivists should follow the descriptive rules con- veyed in Archives, Persoml Papers, and Manu-

scripts: A Catalagcng Manual by Steven Hensen (2nd edition, 1989). AB far as possible, all of the basic descriptive information recorded abut each seriea should be compatible with internal repi tory pdi- ciea and the national descriptive standards which will be outlined in the following chapters. However, practical considerations will usually dictate that the series-level work done during arrangement will have to b comiderably reFmd and placed into a larger context during the fmal description of a group of records.

Once the records are atranged into series they may be arranged on the shelves in any sensible progression. Where m& arrive in a oornpteb and logical order, that order will naturally be main- b i n d . In the abaence of such an order, archivista have developed conventions for the progression of series, beginning with the Dutch rule that general minutes should come first. As David Gracy I1 wrote, series should be asranged in relation to each other "in order of the extent and value of the information within them,"2 This commonly means a progression from the general to the specific, and from policy mak- ing to implementation--such as from executive reco ords Eo progsam records to housekeeping records. Similar prooedures place the records of a central office bfore branches and a director ahead of an assistant direcbr. Within clusters of series at the same Ievel of importance, a chronological arrange rnent may be employed. The thrust of these canven- tions ia to maximize the evidential value of the whole corpus of records in documenting the actual work- ings of an organization.

The ordering of serim within record group has Iong posed a chalIenge to American archivists, who generally eauld not rely on an mtablished framework of European regidxy systems. Yet Holmes r e fersed to it as "the heart of archival work" MUM the archival inventories were essentially reflections of a fmed ordering of the wries on the ~helves.3 Ar- chivjsta believed that just as subgroups had a logical relationship within w o r d groups and file units had an order within series, aa eeriea ~hould be arranged within subgroup or record groups and mlledions. While such ordering ki W u l , its intrinsic importance waa overemphasized; the uee of computers has rendered it even less vital than it had been. Inrhad of relying solely on one fmed wriea order a~ the link between arrangement by provenance and arrangement by series, appropriate computer data base p m

a Gracy, Armmgemmt and Description, 10- Holrnes, "Archival Arrangement," 29.

Arrangement Practices and Procedures 75

mas can provide a whole r a n ~ e of intellectual links m d pawr arrangements, Series can be listed in different orders, and may appear in dwcriptions and Eists as part of more than one s u b p u p or record group. At the same time, the mmplexity and continual change of modern organizations makes it increasingly dificult to fix one appropriate order for records, Even in older colkctiona, the notion that some tm of records are more impartant than others has been questioned by users ranging fmm gene- alogisb h social historians who bypass the minutes and concentrab instead on such "lower" ranked a e ries as case fdes, personnel records, and p m p m records.

The Southside and W e y oollections illustrate many of the problems. The parent Southaide Settle- ment (founded in 1900) i s older than the Tenth Street Settlement (founded in 1903) which it absorbed in 1946. However, placing the records of the latter chm nologically in the middle of the farmer is awkward at best, Placing the Tenth Street files before the records of the larger Settlement has no inherent l a c at all, yet it is the usual procedure. It is conve nient and it preserve^ the separateness of the Tenth Street rmrds, How should the dozens of prmam series documenting the 1920s tn 1968 be ranked? Any ranking that, for example, placd documenting the camp after those documenting the Youth Division would be purely arbitrary, though not in any senw wrong. No rule can tell. the archivist dealing with Bailey's pawn whether to place the business files before the civic files, or the impolrtant Republican Party papers before the more routine Chamber of Commerce files. The archivist ghould concentrate on ensuring that the individual series are coherent and are arranged in some sensible man- ner. At the conclusion of the whoIe prmess, the st& may have been through a wt of records as many as four times-to determine an overall plan of work, to identify groups of fdes, to sort them by series and subseries, and to arrange the series and subseries in relation to each other. (See Figure 7-3.3

Archivkb rearrange records either by break- ing up exiating divisions between series or rearranging fdes within a seri-. As a violation of original order, rearrangement is a controversial practice, though we have seen that Schellenberg endorsed it in the interest of facilitating use. Some archivists have gone beyond the usual consensus that original order should be r e p l a d only where it is incompre

hensible or purely random. They argue that while original order documenting the activity of the organization should be retained, an order documenting merely the activity of file clerks has no value, even if the order is understandable and rational.'

The museum's Education Division can serve as an example. The division might have had an sdmin- istrative semretary who at the end of every year merged reports, programs and other files from all departments within the division into one annual alphabetical file. Thw, under "1959" one might find first the Budget Summary for Adult Programs, fol- lowed by the Budgets for the other departments, then Correspondence for Adult Programs, and then the other departments'Compondence, and so on. The whole file sequence would start over again for the 1% records. In this case the original order hides h t h provenance and activity. To resture the evi- dence of both, the files should be rearranged by department and then alphabetically and chronologically by file type. Thus, all the Adult Programs Department files would be together, and w i t h that subgroup all of the Budgeta would be together and arranged chronologically. This type of arrangement preserves the work of the creators of the records, not their keepers,

Rearrangement of impeded but usable file BYE kme should be very selective. The work is extremely labor intensive. In addition, variant file orders can be created on paper without physical manipulation of the fdes. In most cases of an imperfect order, fdes should only be ream& when the alternative is obvious and the work can be done quickly. Archivists must ensure that tihe new order of the files is based not on serving hypothetical research needs, but on reconstructing the activity that p m d u d the records. Rearch interestg are notariody subjective and changeable. Rearranging materials to suit either one group of individuals (such as academic historians or genealogists) or one type of possible we (such as quantification or legal research) is almost guaranteed to ensure that the records will b usable only by a small group of people or for a short epctn of time, or both.

Arranging FiIe Units and Documents In most cases, the ordering of fde units and

dmuments is a routine clerical activity once the 8e-

rim and subseries have been established, The ordering of folders and bund volumes according to the

See @ally Frank &lee, "Disrespecting Original Or- der," American Archivist 46 (Wintar 1982): 2632.