Bogdan Florin Popovici1 
(Romania)
“YET ANOTHER ONE?!” SOME CONSIDERATIONS ON 
INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL
ABSTRACT
This paper seeks to compare the use of different, but apparently synonym terms, like 
Instantiation, Manifestation or Representation, terms used in different models and pro-
jects in library, archives and digital preservation. For the comparison, we reviewed the 
definitions and examples in several authoritative documents, like PREMIS, FRBR, Re-
cords in Contexts etc. This exercise revealed that similar words may have different us-
age in different context, which make it hard to use one term to communicate the same 
meaning for everybody. The conclusion is that, at least for the time being, Records In 
Context model of archival description needed to use a new term and to define its scope 
of use within the framework of the descriptive model, since it present specific features 
and aim to be general, carrier-neutral. 
Keywords: Records in contexts, PREMIS, digital preservation, archival description, FRBR, 
PAAST
1 INTRODUCTION
During working meetings of the Experts Group for Archival Description (EGAD), under 
the auspices of the International Council on Archives, in the process of identifying the 
relevant entities, several existing models have been analysed. For one specific case, it 
was agreed to use a “new” term (i.e. “not used in other archival-related standard”), as 
it was easier to model a new entity that fits to the needs of the model than to create 
potential confusions with other models’ entities. However, I was inclined ever since to 
have a closer review of such other entities, similar with the one used in “Records in Con-
texts” model (hereafter, RiC). 
Despite the fact a certain amount of time was devoted to the issue of Instantiation, a 
question that bothered me was if it was really necessary to employ a new term to label 
this entity, while some other terms, apparently similar, already existed. Manifestation 
(in FRBR or PAAST) or representation (in PREMIS) seem quite similar with Instantiation. 
Hence, the purpose of this paper is to comparatively examine these terms and check 
their suitability.
1 Bogdan-Florin Popovici graduated the Faculty of Archival Science In Bucharest (Romania) (1999). He 
holds a Ph.D. in History (2004). He attended several training programs in Europe (France, Italy). He works 
with the National Archives of Romania, Brasov County Division (since 2009), as archivist and, since 2013,  
as head of Division. He has been actively involved with ICA/Committee on Best Practices and Standard, 
respectively Experts Group for Archival Description (2008-). He participated at several international 
professional meetings (Slovenia, Italy, Oman, Russia, Hungary, Mexico, South Koreea, Spain), both as 
a speaker and as a professional trainer. He is also involved in translating professional standards into 
Romanian. Contact: bogdan.popovici@arhivelenationale.ro.
48 “YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici
2 THE CASE
Despite having as a condition of its existence the attribute of “fixity”, a record passes 
through a lot of changes in its being, mainly of context, but sometime also of form and 
appearance. It is, as the Australian archival scholars called it, the Records Continuum. 
Sometime, these transformations do not change significantly the record, one observer 
being still able to call it “the same” record2. In other situation though, such a change may 
lead to a new record, related with the initial one. From the perspective of an archivist, 
describing records that encompass transformations may mean repetition of bits of de-
scription, that are shared by each instance. 
Here are some examples. We may have:
• a charter from 15th century, which is described by archivist. We also have a microfilm 
copy of this charter, that needs to be described. No matter if we consider that it is “the 
same” record or not, some descriptive values are identical (for example information-
al content), while others are not (e.g., carrier). 
• A historian’s personal paper. Among the documents, there is a transcription of a 18th 
century record. When describing it, we may re-use parts of the description of the 
original document for describing the focus of the historian’s record. 
• a 14th century record which embeds an extended transumptum of an older charter. 
When compiling a document chronological edition, that transumptum will be used 
as a placeholder for the original document, which is missing today; the transumptum 
will be the document, in its abridged form. 
• a letter, that was in the same time produced as paper record and as a DOCX file, which, 
for preservation reasons, was then converted to PDF. The informational content (the 
message) is the same, but each instance has its own characteristics. 
• a paper original, which is digitized, then the paper original is destroyed. Within that 
organisation, it is considered “a copy in the form of original”; and so, we may have a 
nice digital record from 1950... 
Many similar cases are met in archival practice. In all the examples, there is “something” 
which is not the original document, but a different entity, that may be called a “copy”, a 
“reproduction”, a “variant”, a “transformation” etc. Describing each one of these implies 
inevitably to use some shared elements (form, appearance, content), in addition to spe-
cific elements of each of that “something”. In some cases, the two related documents 
may be considered “the same”, while for others they are regarded as being different; 
the odd situation is that all sides are right, depending upon the context. 
These cases are not new and in practice they were found, and descriptive method of 
referencing was used for long time. While referencing may be used within digital tools 
for archival description, the question is whether that “something” does not reveal the 
existence of a new entity, tight connected with record, but with its own attributes. 
The Records in Context Conceptual Model calls this “something” an instantiation and 
considers it as one entity of archival description. 
2 For another contribution to the topic see our paper presented in Kazan (Russia) and published in Roma-
nia (Popovici, 2019).
49“YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici
3 THE “INSTANTIATION”
“Records in Contexts”, as a conceptual model for archival description, has not yet 
been released formally as a “definitive” version (EGAD, 2019). Despite that, for the 
purposes of present paper, we shall analyse the concepts considering the present 
status of development. 
Traditionally, “record” or similar terms were used to define “archival materials”, but RiC 
splits it into several different entities. A first separation is within the archival material 
types, where RiC individualizes record, component of a record and aggregation of re-
cords as Record, Record Component and Records Set entities. Archival material as such is 
labeled Records Resource. A second differentiation is made between Records Resource, 
as intellectual entity, and its physical presence, as Instantiation. 
The current definition for Instantiation in RiC is “The inscription of information made by 
an Agent on a physical carrier in any persistent, recoverable form as a means of commu-
nicating information through time and space”, while record is “Information inscribed at 
least once by any method on any physical carrier in any persistent, recoverable form by 
an Agent in the course of life or work Activity.” 
As resulting from the definition, an instantiation is a sine qua non condition for the birth 
of record; if no instantiation exists, we cannot talk of a record (e.g., a radio broadcast or 
a “signature” on water surface). However, the original instantiation may disappear in 
time, and record may only be preserved either a copy or a notice (e.g., transumptum). It 
means that, although the record does not exist in its original instantiation, the message 
in context may be considered to survive through its new, derived, instantiations, which 
is a new record, but functionally equivalent. 
A record may have multiple instantiations. They can be simultaneous (e.g., a scanned 
record, generated in one step as TIFF and JPG), or successive in time (a paper record, mi-
crofilmed, then digitized). The latter re-instantiated record, depending on the context, 
can be considered the same (functionally equivalent of the original instantiation) or a 
new record. For example, the microfilm in the Archives is considered as a surrogate for 
the original, and many users find this is quite “the same” record (while it is obvious it is 
a new record, with different date and carrier). However, a police officer would hardly 
consider a photocopy of the driver licence as being the original. 
In case of Records Set (e.g., file, series or even fonds), the original instantiation is a sum 
of each component record’s instantiation. But in the case of re-instantiations, a reshape 
may occur. For instance, a file is a collection of records on paper; despite being treated 
as a whole, each of them keeps also individuality of carrier, form etc. After a digitisation, 
all the records in the physical file may be assembled in one PDF file, in which case their 
individuality as items can be disputable. Reciprocally, a ledger (one physical item) can 
be re-instantiated digitally as a JPG file for each page. The new instantiation may be, 
therefore, a fragment or a sum of previous instantiations. 
By their nature, Instantiations are the main target of preservation. Either analogue or 
digital, Instantiations implies a carrier and/or a form of encoding that may require an 
intermediation device (from paper, vinyl disk and magnetic tape to digital file format). 
This aspect is important within the comparison with other models and explains why 
digital preservation focuses on instantiation related issues. 
Existence of an intellectual entity and its instantiation were contemplated in other 
models too. The labels used were different. In the following we shall examine some 
of them. 
50 “YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici
4 THE “REPRESENTATION”
PREMIS (Preservation Metadata: Implementation Strategies) is an international stand-
ard seeking to offer support for practical implementation of metadata in digital preser-
vation processes. Started in 2003, the project has become one of the basic intellectual 
tools for digital preservation worldwide. 
The PREMIS Data Model defines 4 entities: Object, Right Statement, Agent and Event. 
For the purposes of the present paper, the entity of interest is Object, which is defined 
as “a discrete unit of information, subject to digital preservation”. Object has 4 subcat-
egories: Intellectual Entity, Representation, File or Bitstream. The Intellectual Entity is 
defined as “distinct intellectual or artistic creation that is considered relevant to a des-
ignated community in the context of digital preservation”. A Representation is „the set 
of files, including structural metadata, needed for a complete rendition of an Intellec-
tual Entity”. One Intellectual Entity can have one or multiple Representations; in fact, it 
will always have at least one Representation. A File is considered a „named and ordered 
sequence of bytes that is known to an operating system” and it designates what is com-
monly understood as “file” in a computer systems, while Bitstream goes deeper in the 
structure of a File, indicating “the contiguous or non-contiguous data within a file that 
has meaningful common properties for preservation purposes” (PREMIS Data Diction-
ary for Preservation Metadata, 2015, pp. 7-8). The conceptual view is represented in the 
quoted resource like this: 
Figure 1 Subcategories of Object and their relations in PREMIS (apud PREMIS)
In a practical example, the image of a mountain would be the Intellectual Entity, as an intel-
lectual creation and unit of information. The image as a digital photograph is saved on the 
card in 2 files, in format JPG and RAW, both of them being Representations of the picture. 
Beside digital Representation, in the latest version of PREMIS (v.3), it is accepted that an 
Intellectual Entity may also have non-digital Representations. In the examples above, 
a charter, a transcription, a transumptum, the letter printed are all Representations of 
their original record (which are the Intellectual Entities), either physical or digital. More-
over, as indicated in PREMIS Data Dictionary, the Intellectual Entity may not be only re-
garded at a granular level, but also encompassing large components, like archival fonds 
or series. While the inclusion of non-digital Representation is rather non-consistent just 
yet (for instance, it is clear that non-digital Representation may not include Files, as de-
picted in the Figure 1 above), it seems that Representation, in its core characteristics, is 
similar to Instantiation. 
51“YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici
5 THE “MANIFESTATION”
One term heavily used in cases of multiplication of the content in various instances is 
manifestation. Despite the common term, the meaning seems inconsistent among var-
ious documents. 
A). FRBR
Functional Requirements for Bibliographic Resources is a user-centred conceptual mod-
el for describing objects of interest for librarians/users of bibliographic information 
(IFLA, 1998, p. 2). Unlike PREMIS, that embraces physical entities later in its develop-
ment, FRMR was addressing physical and digital as a core option from the beginning.
The model comprises several entities, from which four of them in the first group (Prod-
ucts of Intellectual & Artistic Endeavor) are relevant for the present paper. Work is de-
fined as “a distinct intellectual or artistic creation”. Expression is “the realisation of a 
work, wither intellectual or artistic”, which may take a variety of forms (textual, sound, 
musical notation, images or a combination of them). These two entities are purely ab-
stract. Manifestation represents the physical embodiment of an expression of a work 
(that is, information put in a container or on a carrier). It is normally associated with a 
particular medium, such as a printed volume, photographic print or film reel. Item is the 
copy (exemplar) of a manifestation, that is, a physical instance. A manifestation may 
have more than one item, but one item can only have one manifestation. (IFLA, 1998, 
pp. 17-24). “A manifestation represents all the physical objects that bear the same char-
acteristics of intellectual content and physical form. In actuality, a manifestation is itself 
an abstract entity, but describes and represents physical entities, that is all the items 
that have the same content and carrier.” (Tillett, 2003, p. 10)
B). DIGITAL PRESERVATION — BROWN
Reflecting the findings of several projects and of the practical activity in the field, Adrian 
Brown approaches, in an influential book, some practical issues in digital preservation. 
Due to the methodologies used in this process, it is inevitable an analysis of the concepts 
associated with various “embodiment” of a record. 
Describing processes and entities, A. Brown uses several terms, as “manifestation”, “ver-
sion” or “representation”. Manifestation is defined as “specific data object that instanti-
ates an information object. Multiple manifestations can exist for any given information 
object” (Brown, 2013, p. XIV). Later, the author emphasizes a very important feature: 
“it is critical to distinguish between manifestations, being technical representation of 
the same information object and different editorial version of an information object” 
(p. 216). The example given covers different ways that the same information object3 (a 
book) can be physically stored:
3 Defined, in the book, as in OAIS, as “a conceptual object of preservation… realized as meaningful infor-
mation by interpreting a data object through its associated representation information”. 
52 “YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici
Figure 2 Manifestations (apud Brown)
Adrian Brown also scrutinizes the origins of term “manifestation” within the realm of 
digital preservation. His conclusion is that term “almost certainly evolved independent-
ly within a number of different initiatives” (Brown, 2013, p. 215) and he gives no indica-
tion that the term might have been borrowed from library conceptual model. 
C). DIGITAL PRESERVATION — INTERPARES PAAST
Another relevant document for using the term “manifestation” is PAAST, a product re-
sulted from last InterPARES Trust project. Preservation as a Service for Trust (PaaST) 
presents functional and data requirements for digital preservation (Interpares_Trust, 
2017). In this project usage, Manifestation is “A concrete, physical implementation of an 
Intellectual Entity capable of communicating information as originally intended”, while 
an Intellectual Entity is “an artifact that is intended to communicate information”, a re-
cord being one type of Intellectual Entity. 
In PAAST model, an Intellectual Entity is digitally encoding its specification in an enti-
ty called Binary Encoding. The Binary Encoding has 2 type of parts: Digital Component 
(that is, “an ordered string of bits that encodes numbers, characters, signs, symbols, 
sounds, images or other graphics in a digital form that is both suitable for storage and 
appropriate to instantiate an Intellectual Entity”) and Manifestation. This perspective 
is consistent with previous InterPARES findings, which discussed about the stored and 
manifested records: “Stored record is constituted of the digital component(s) used in 
re-producing it, which comprise the data to be processed in order to manifest the re-
cord (content data and form data) and the rules for processing the data, including those 
enabling variations (composition data)” and “manifested record is the visualization or 
instantiation of the record in a form suitable for presentation to a person or a system” 
(Duranti & Thibodeau, 2006, p. 39), (Duranti, 2016, p. 8).
Manifestation, from PAAST view, enables Intellectual Entity to be instantiated. For ex-
ample, a record (Intellectual Entity) is digitally encoded as a PDF file (record stored); 
manifestation is what would be rendered on/readable from a screen by a human reader 
(that is, the message conveyed reached to interested party)4. Therefore, a PAAST Man-
ifestation is a physical materialization of an Intellectual Entity, which is not stored and 
preserved, but exists on an output device (Interpares_Trust, 2017, p. 23)
4 This is an intentional simplification for our paper; PAAST specifies that Manifestation may be under the 
form of a Runtime version (when the manifestation is intended for a machine) or of a rendering (when 
the IE is output in a form accessible to humans). 
53“YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici
Figure 3 Composition of an Intellectual Entity (apud PAAST)
6 SOME CONCLUSIONS
Reviewing some of the meaning of terms describing physical embodiment of an intel-
lectual entity, a first remark can be the semantic proximity. “Manifestation” is used in 
several documents, and it seems it has similar, but not identical meanings. On the oth-
er hand, visible mainly at A. Brown, manifestation can both “represents” or “instanti-
ates” a piece of information, employing the same wording as in RiC (Instantiation) or 
PREMIS (Representation). This emphasizes again that the meanings for Manifestation, 
Representation and Instantiation are close, though not necessarily fully overlapping. 
As a scope, FRBR is the closest to RiC, since it aims to encompass all “information objects” 
without a dedicated view on physical or digital objects. PREMIS is (tentatively) close but, 
in our opinion, the presence of non-digital Representation is not yet very elaborated, 
and it is rather peripherical than central to the model (which focus on digital preserva-
tion). On the other hand, FRBR has certain features that makes it less suitable for use in 
archival description and so, in using its entities to describe archival ones. Usually, archi-
vists’ interest towards abstract entities, like Work or Expression may be considered rath-
er low, since record is rarely regarded as an intellectual or artistic product, but rather an 
instrument or a by-product of a practical activity (as definitions cited above indicate). 
As about Manifestation, as it was noticed, “As entity, the manifestation is the set of all 
physical objects that shares…” (Tillett, 2003, p. 10). But creating many copies of a record 
is barely the norm. Where many copies exist, they may fulfil different functions in dif-
ferent contexts, and this makes them unique, even if the information content and phys-
ical form may be identical. A similar distinction was made in PREMIS, where the term 
“Representation” was used specifically to avoid the term “manifestation” as it is used 
in the FRBR. “In the PREMIS model, a Representation is a single instance of an Intellec-
tual Entity held in a preservation repository”. (PREMIS Data Dictionary for Preservation 
Metadata, 2015, p. 12), unlike FRBR. This brings PREMIS_Representation very close to 
RiC_Instantation.
54 “YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici
The term Manifestation used in digital preservation seems to have different meanings 
for different authors. For A. Brown, according to the definition, Manifestation is data ob-
ject and it is synonym with Representation. The same word, used in PAAST refers to the 
actual communication of information, while data object is the Binary Encodings. Using 
the term Manifestation in RiC would implicitly lead to misunderstandings, since there is 
no apparent consensus of using the term.
PREMIS_Representation, and Manifestation as it is used by A. Brown seem to be to very 
close to RiC_Instantiation. It is about the mean used to inscribe information for con-
veying it, not about the information conveyed to the user (which is abstract before 
inscribed). Any Intellectual Entity can have multiple representation/manifestation/in-
stantiations. Brown underline the fact that manifestations are “technical representa-
tion of the same information object” and they are different than editorial versions of the 
same information object (Brown, 2013, p. 215), with the example of a cropped photo, 
which is regarded as a new conceptual object). In similar way, a record instantiated can-
not be changed or modified, so a re-instantiation is a new entity, which can (or not) be 
considered as functional equivalent with the original.
It needs to be emphasized that RiC_Instantiation aims to be carrier/medium independ-
ent; it needs to describe both analogic and digital inscription of records, while PREMIS_
Representation/Manifestation were shaped for digital realm. Expanding the under-
standing of PREMIS_Representation open to path to synonymity to RiC_Instantiation, 
but in this moment it is not clear if there are some fine differences between RiC/Mani-
festation or Representation or not. So, for now, it may be a good option to use new term, 
in order to satisfy the description needs. But very likely, in the future, a convergence of 
meaning will lead to a unification of concepts. 
REFERENCE LIST
(EGAD), E. G. (2019). Records in Contexts (v.0.2). Preluat de pe https://www.ica.org/
en/public-release-of-records-in-contexts-ontology-v01-and-records-in-con-
texts-conceptual-model-v02
Bogdan-Florin, P. (2019). „Cât de bine poți cunoaște un document?” Despre dimensiuni-
le fizice si intelectuale ale documentelor arhivistice . În Szekély Százodok. Emlék-
könyv Pál-Antal Sándor 80 születésnapjára (pg. 410-424). Marosvásárhelyi, .
Brown, A. (2013). Practical Digital Preservation. A how-to guide for organisations of any 
size. London: Facet Publishing.
Duranti, L. (2016). The Concept of Record in the Digital Environment (presentation).
Duranti, L., & Thibodeau, K. (2006). The Concept of Record in Interactive, Experiential 
and Dynamic Environments: the View of InterPARES. Archival Science, 6(1), 13–68.
IFLA. (1998). Functional Requirements for Bibliographic Records. Munchen. Preluat de 
pe https://www.ifla.org/files/assets/cataloguing/frbr/frbr.pdf
Interpares_Trust. (2017). Preservation as a Service for Trust (PaaST). Functional and Data 
Requirements for Digital Preservation. Preluat de pe https://interparestrust.org/
assets/public/dissemination/PreservationasaServiceforTrust1_0.pdf
PREMIS Data Dictionary for Preservation Metadata. (2015). Preluat de pe http://www.
loc.gov/standards/premis
Tillett, B. B. (2003). The FRBR Model (Functional Requirements for Bibliographic Re-
cords). Preluat de pe https://www.loc.gov/catdir/cpso/frbreng.pdf
55“YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici
FURTHER REFERENCES:
Gill, T. (2004). Building semantic bridges between museums, libraries and archives: The 
CIDOC Conceptual Reference Model. First Monday, 9(5–3 May).
Henttonen, P., & Kilkki, J. (2017). “Records in contexts” and the Finnish conceptual model 
for archival description. Letonica, 36, 60–71.
Howarth, L. C. (2012). FRBR and Linked Data: Connecting FRBR and Linked Data. Cata-
loging & Classification Quarterly. https://doi.org/10.1080/01639374.2012.680835
Le Bø euf, P. (2012). Modeling Rare and Unique Documents: Using FRBROO/CIDOC CRM. 
Journal of Archival Organization, 10(2), 96–106. https://doi.org/10.1080/1533274
8.2012.709164
Llanes-Padrón, D., & Pastor-Sánchez, J.-A. (2017). Records in contexts: The road of 
archives to semantic interoperability. Program, 51(4), 387–405. https://doi.
org/10.1108/PROG-03-2017-0021
Riva, P., Doerr, M., & Žumer, M. (2009). FRBRoo: Enabling a common view of information 
from memory institutions. ICBC, 38(2), 30–34.
Thurman, A. C. (2007). FRBR and archival materials. In A. G. Taylor (Eds.), Understanding FRBR: 
what it is and how it will affect our retrieval tools (ss. 97–102). Libraries Unlimited.
56 “YET ANOTHER ONE?!” SOME CONSIDERATIONS ON INSTANTIATION ENTITY IN ‘RECORDS IN CONTEXTS’ MODEL Bogdan Florin Popovici