THE NEED OF ‘INFORMATION ANALYSIS’ FOR
INFORMATION SYSTEMS AND OUTLINE OF A
HERMENEUTIC APPROACH TO IT
Sufen Wang
Glorious Sun School of Business & Management, Donghua University, Shanghai ,200051, China
Junkang Feng
University of Paisley UK, High Street, Paisley PA1 2BE, UK
Binyong Tang
Glorious Sun School of Business & Management, Donghua University, Shanghai ,200051, China
Keywords: Information Systems, Requirements Analysis, Hermeneutics.
Abstract: Requirement analysis for information systems development (ISD) results in a specification that represents a
central reference point for subsequent stages of the development. But this stage is seen characterized by
informality and uncertainty. One essential element in this is how the information that is required by the
users in a domain is identified and formulated. In this paper, we give a summary on how well-known
information systems methodologies handle it, and argue that an information analysis would seem to be
needed. Such an analysis would require a certain approach, for which we suggest adopting Hermeneutics.
We describe how Hermeneutics might enable us to look at the mechanism whereby information is created
and information flow takes place in the context of information systems.
1 INTRODUCTION
Two basic assumptions seem to have been made in
information system (IS) literature. The first one is
that an information system is a means by which
information is provided. The second is that an
information system can only handle data, and data
carry information. Based upon these two
assumptions, approaches to IS development should
emphasize analysis, and take the finding of
requirements for an organization as its central task
(Flynn 98, Mingers 92 and Loucopolous 92). At the
centre of the requirements determination and
analysis is how required information is identified
and formulated. But this phase is also known as
‘characterized by informality and uncertainty’
(Loucopoulos and Zicari 92), and ‘the least well-
defined phase in the systems development process’
(Flynn 98).
In this paper, we present a summary of a
literature review and then suggest that an
'information analysis' stage be included in ISD, and
a mechanism for this stage be developed. This paper
is organised as follows. In section 2 we give the
summary. In section 3 we outline a Hermeneutics
foundation for a mechanism that can be used for
such a stage. We conclude this paper with a
summary.
2 THE NEED FOR
INFOTRMATION ANALYSIS
FOR ISD
We look at how well-known IS methodologies
handle the problem of information requirements in
terms of three aspects, namely
The content of IS requirements by the user
506
Wang S., Feng J. and Tang B. (2007).
THE NEED OF ‘INFORMATION ANALYSIS’ FOR INFORMATION SYSTEMS AND OUTLINE OF A HERMENEUTIC APPROACH TO IT.
In Proceedings of the Ninth International Conference on Enterprise Information Systems - ISAS, pages 506-511
DOI: 10.5220/0002347505060511
Copyright
c
SciTePress
How required information is identified,
and
How data that an IS will actually store and
process are identified.
It would seem that the main problem in the
aspect of ‘the content of IS requirements by the user’
is that the requirements are modeled by a variety of
conceptual models (normally a semantic data model
and a process model of some kind, for instance in
DFD), and the information required by the agent or
user is not captured by means of some workable
definition, concrete format and systematic
procedure.
In the aspect of ‘how required information is
identified’, the main approach is that a variety of
information related things, such as information
category, information set, information flow, are
identified by looking at whether they are necessary
for an activity or process, or whether they are input
or output of it, or processed by it. The main problem
with this approach is that even all of them put
together might still not be sufficient for an activity.
Moreover these information related things are not
necessarily the information that the agent really
requires.
In the aspect of ‘how data that an IS will actually
process are identified’, it would seem that the main
problems are, first of all, a data schema is not
derived from required information captured in a
concrete format. As information and data are seen
essentially the same thing despite different terms
being used, to construct a data structure is a matter
of data modeling through data consolidation,
specialization, aggregation or step-wise refinement.
That is to say, a data structure is not always
explicitly derived from required information.
Secondly as a consequence of the first one, these
approaches do not address the issue of how data in a
machine bear information that can be derived by a
user from them. There is no clearly defined concept
of and mechanism for looking at how data are
structured to bear required information. Batini et al.
(92) raises the question of the 'information content'
of a data schema, but does not seem to have
provided an adequate solution. Moreover this
question is not well acknowledged in the literature.
To address these problems just identified, an
'information analysis' stage would seem desirable,
which should be inserted between problem
appreciation and conventional data modelling and
process (or function) modelling. This stage would
require a mechanism, which should be based upon a
formal, useful and systematic treatment of the
concept information. We suggest adopting
Philosophical Hermeneutics as an approach to it
with which how information is created and
information flow takes place in the context of
information systems can be looked at.
3 A HERMENEUTIC APPROACH
TO THE PROBLEM
Boland (1987) points out that how well the nature of
information is addressed has profound impact to all
aspects of information systems – the research,
development and use of information systems, both in
theory and in practice.
We suggest using results of advances in the
research of ‘information philosophy’ in recent years
thereby to formulate a new perspective for obtaining
a better understanding of the discipline of
information systems and for investigating the
essence of information. That is, to look at the
problem of information that is meaningful to its
creator and its user by drawing on philosophical
thinking on the notion of ‘existence’. We believe
that what we call meaningful information has to be
approached from the point of view of the inter-
relationship between information, data and meaning;
and the inter-relationship can be understood through
a process of ‘inter-exchange’. That is, in the context
of IS, information is normally carried (borne) by
data organized in some way, which results in
meaning for the receiver, and then the meaning that
the receiver creates in her/his mind, which is
attributable to the information that she/he has
received, might produce new information if some
reduction in uncertainty takes place; and the new
information may be borne by new data. This would
be the start of a new cycle. Such a process never
ends. We also note that this takes place against and
be influenced by the real world, and the latter is the
ultimate source of information. We call such a
process ‘information-data-meaning-information-
data…’ cycle. We observe that such a process does
not seem to have been well understood.
3.1 Why is the ‘Information-Data-
Meaning-Information-Data…’
Cycle a Problem of Hermeneutics?
Hermeneutics is the study of interpretation.
Hermeneutics emerged as a concern with
interpreting ancient religious texts and has evolved
to address the general problem of how we give
meaning to what is unfamiliar and alien (Boland,
THE NEED OF ‘INFORMATION ANALYSIS? FOR INFORMATION SYSTEMS AND OUTLINE OF A
HERMENEUTIC APPROACH TO IT
507
1991). Gadamer argues that the problem of
interpretation is fundamental to our everyday
activity ------ also in the business world.
(Hirschheim et al., 1995p.152)
Information in the context of IS consists of that
of organizational environment, that of the operation
of a business, that regarding how the user uses an IS,
and so on. All these are normally represented by
various types of data, such as user requirements,
data models, data about the system, input/output of
computer systems, data created in business
operations. These are all data from which we want to
obtain meaning through interpretation. Thus these
data are in the position of the target, i.e., ‘text’ in
Hermeneutics.
These data, information and meaning are in a
state of co-existence. Information is borne by data,
and meaning is created due to reception of
information through looking at data.
The process of the ‘information-data-meaning-
information-data…’ is equivalent to that of reading
and writing, listening and speaking between two
people. Thus, the inter-relationship between
information, data and meaning through the
‘information-data-meaning-information-data…’
cycle embodies social exchanges, through which
people obtain their understanding of the world in
which they are, and in turn through the understand
they obtain their existence. At the same time,
through social exchanges the social world we are in
changes as well.
The idea of tackling data in IS with
Hermeneutics has appeared in the literature, for
example, Gadamer was referred to. But the relevant
work shown in the literature did not go the details,
for example, why and how Gadamer can be applied
to IS was not elaborated. Besides, as Hirschheim et
al. 1995, p.154 point out, the importance of
philosophical hermeneutics including the ideas of
hermeneutic cycle, horizon, and bracketing for rule-
based data modeling is far reaching, and therefore
further investigation down this line would seem
desirable and possible.
3.2 The General Process of
‘Information Forming’ Mechanism
The classical Hermeneutics put forward by
Schleiermacher and Dilthey, and the General
Hermeneutics theory by Betti, Hirsch and so on
take Hermeneutics as a common methodology of
humanity, and also a epistemology (Zhang, 1985).
On the contrary, philosophical hermeneutics by
Heidegger (1962) and Gadamer (1975, 1976) take
that hermeneutic is concerned with the most
fundamental problem of human existence, a problem
of ontology. Ricoeur maintains that the existential
ontology can only be achieved through the
investigation of methodologies, and layers of
epistemology (Zhang, 1985). This way, we not only
retain the usefulness of the methodological studies
of a long history and the unique insight of its
epistemological aspect, but also take the notion of
reflection of epistemology to the level of ontology.
When we consider information systems, no matter
whether they are seen as technological systems with
social consequences, or as social systems that are
technically and technologically implementedthey
have a social aspect along with a technical and
technological one. Therefore in order to draw on the
vast resources of research on information systems
based upon the view that takes information systems
as technological systems, we adopt Ricoeu’s
Hermeneutics as the theoretical foundation for our
investigation into the mechanism that enables the
formation of information and information flow
within the context of information systems.
Ricoeu combines ontological Hermeneutics with
methodological and epistemological Hermeneutics
through linking Hermeneutics with the text theory.
The general process of information and
information flow formation (see Fig.1) may be seen
as having three stages, namely the Semantics Layer,
Reflection Layer, and Ontological Layer. Each of
the layers is connected with the ‘text’ (i.e., data) of
the information system. The transformations
between the three layers embody those between
objective meaning (in the sense of being
independent of the receiver of information), inter-
subjective meaning and subjective meaning.
Figure 1: The general process of information.
3.3 The Analysis of Various Elements
in the ‘Information Forming’
Process
3.3.1 Data Analysis
Data in information systems are read and interpreted
as texts. There are various definitions about data that
can be found in the literature (Hirschheim et al.,
ICEIS 2007 - International Conference on Enterprise Information Systems
508
1995p.12; Mingers, 1995; Floridi, 1999). We give
data here slightly different characteristics from those
that appear in more ‘general’ research of information
systems.
We think that data links information and
meaning, which enables the communication between
people. Through communication, people acquire
self-understanding. Thus data should have the
following characteristics:
(1) Data are fixed life expressions by being written.
They have “dual-meaning” or “multi-meaning”.
There are literal meaning, sender’s meaning, hidden
and latent meaning produced by various factors,
such as the multiple traits of the literal meaning, the
knowledge background and psychological factors of
the sender and so on.
(2) There is a dialectic relation between the sender’s
meaning and the meaning that may be seen as
inherent to the data. They are inter-dependent, which
shows the relation between data and information.
The structure of data determines the average
quantity of information that data can carry (Dretske
81). The stipulation or protocols of encoding about
signs used in data determines which events data can
express and therefore what particular information a
data item actually carries. For example, we might
flip a coin in order to decide whether Miss Wang
would be on duty today. We may pre-arrange that if
the coin were ‘face up’, then Miss Wang would be
on duty. Therefore, the data ‘the coin is face up’
would carry the information that Miss Wang is on
duty today.
(3) The meaning created through the information
that is carried by data and the importance of data are
derivate from the dialectic relation between data and
its receiver. The dialectic relation embodies the
relation between information and meaning, and the
relation is realized by jointing data and meaning.
The meaning and the importance are achieved
through the increase of the receiver’s understanding
of the world under the influence of the information
carried by data, which is characterized by the
reduction in the uncertainty of the receiver’s
understanding of the world. If such reduction in the
uncertainty does not take place, then the information
carried by data would be meaningless to the user of
the data. For example, if I already knew that Miss
Wang is on duty today, then the data that ‘the coin is
face up’ would be of no meaning to me in the sense
that it gives me nothing except perhaps annoying
me.
(4) Data is not limited by their direct references; data
enable people to enter a possible world from a given
world, i.e., the data world. The references here are
not only positivist meaning or scientific meaning in
their general sense, but also metaphoric meaning.
For example, the data ‘the coin is face up’ carries the
information that Miss Wang is on duty today, then it
can further be known that Miss Wang is not at home
today.
3.3.2 Semantic Analysis - The Analysis of
Information Content
The analysis of the information content of data,
through interpreting the data, we can obtain
objective information content carried by the data.
The objective information content is taken as the
meaning that the sender of the data wishes the data
to carry. So ‘objective’ here means being
independent of the receiver of the data. Data may
have various meanings. For example, the data that
‘the coin is face up’ has the literal meaning that the
coin is face up. Under a pre-defined encoding rule
for communication, the data also carry the
information that the sender of the data want to
express, namely ‘Miss Wang is on duty today’.
There could be other implied information content
such as ‘Miss Wang is away today’.
Literal meaning is the direct and basic meaning,
and the others are called indirect, second or
metaphoric meaning. These indirect meanings are
nested within the direct meaning. This is similar in a
way to information nesting (Dretske81).
We begin with interpreting data that have multi-
stipulations. But every kind of interpretation is based
on its own frame of reference in order to seek
agreement with the rich and multi-vocal meanings of
data. The interpretation process of data is illustrated
in Figure 2.
Figure 2: Interpretation procedure.
The analysis of information content, namely data
interpretation, must follow two principles of the
hermeneutic cycle. The first principle is concerned
with the relationship between parts and the whole
and the second principle between understanding and
experience. The relationship between parts and the
whole is this: parts can only be ones in a whole
rather than being independent of the whole; We can
try and understand the whole through looking at the
THE NEED OF ‘INFORMATION ANALYSIS? FOR INFORMATION SYSTEMS AND OUTLINE OF A
HERMENEUTIC APPROACH TO IT
509
parts; But this understanding is approached by
looking at how the parts behave within the whole.
The relationship between understanding and
experience is this: on the one hand data wait for an
interpreter to be interpreted, and on the other hand
the interpreter can only understands those that his
experience allows him to see. This is a process
where known and experienced things are used as a
tool for the interpreter to reveal the existence of the
unknown. In a word, this is concerned with how to
connect the known with the unknown, which is
much larger than the known. In addition, it is the
background that gives meaning to those that are
known.
3.3.3 Reflection Layer
The information forming process embodies the
communication between people by means of the
inter-relationship of data, information and meaning.
Its goal is for people to achieve understanding of
themselves by communicating one another.
Thus our interpretation of data is not just the
understanding of the information content that is
carried by the data, but also the meaning of the
sender of the data. The purpose of this is, through
understanding the sender’s meaning, to ascertain
what world we ourselves are in, and make sure of
what ‘I’ am, and what I should do. This is self-
understanding, to achieve which there has to be a
process of reflection.
Reflection is of course self-reflection, and not a
concrete reflection on a particular event. Reflection
is a process of transforming the ‘otherness’ of the
data into an ‘utterance event’ for me. The receiver’s
‘utterance event’ is a new event, that is, it is not the
repetition of the ‘‘utterance event’ that created the
data in the first place, but is new creation according
to the requirement of ‘speaking’. This way, the
interpretation of reflection is completed. Thus, self-
understanding is realized through reflection.
Reading links two incidents of speaking: data as
utterance, and reading as new utterance (See Figure
3). Ricoeur makes use of Gardmer’s ‘fusion of
horizons’ to refer to the widening of the
understanding of the subject after she/he has entered
the world of data.
Reflection process is completed through reading
data and conversing with data, and reading through
‘fusion of horizons’ and game-playing.
3.3.4 Ontological Layer
After reflection, self-understanding comes into being
according to the form in which it can exist, and it
creates new data. This is not an end, but the
beginning of a new cycle. This process of
information and information flow formation
constitute a basis of exchange between human being.
3.4 Concluding Remarks
Through semantic interpretation of the semantic
layer, the receiver obtains the information content of
the data sent by the sender. Much of the information
content exists in the form of being implied and
implicit, through obtaining which the receiver
obtains her/his understanding of the sender. Through
assimilation via reflection, the receiver strives to
find the way to further understanding her/himself,
namely to make something ‘alien’ to be of his/her
own. On the ontological layer, the receiver expresses
his/her own utterance with new data. Through such a
never-ending cycle, human exchange is achieved,
through which we increasingly under ourselves (see
Figure 4).
We identified the three problems in the previous
section concerning requirements of IS, namely,
The content of IS requirements by the user,
How required information is identified,
and
How data that an IS will actually process
are identified.
These can now be looked at within such a never-
ending Hermeneutic cycle. The content of
information requirement from the point of the view
of the user would now be what is needed for her/him
to understand her/himself in the context of using an
IS to approach and complete her/his tasks and the
meanings that are subsequently produced. The
required information should be identified through
the stages of semantic understanding, reflection and
ontological realization. Finally, the data that an IS
processes should be among the original set of data
and the new data. To work out the details of how a
mechanism for analyzing information and
information flow within the context of IS would
require much more work and it is therefore beyond
the scope of this paper.
4 SUMMARY
In this paper we have presented a summary of a
literature review to support our observation that an
‘information analysis’ stage for information systems
development would seem to be needed, in which the
information that an agent in a domain requires is
identified and formulated. Such a stage would need a
ICEIS 2007 - International Conference on Enterprise Information Systems
510
mechanism. To develop such a mechanism, we
suggested using Hermeneutic as the foundations and
we outlined a Hermeneutic approach to information
in information systems.
REFERENCES
Batini, C., Ceri, S., and Navathe, S.B. Conceptual
Database Design: An Eentity-relationship Approach
Redwood City in Calif, Benjamin/Cummings, 1992.
Boland, R.J.Jr.1987. The In-formation of Information
System. In Critical Issues in Information Systems
Research: edt. R.J. Boland, Jr. and R.A. Hirschheim,
John Wiley & Scons Ltd.
Boland, R.J.Jr.1991. Information System Use as a
Hermeneutic Process. In Information Systems
Research: Contemporary Approaches & Emergent
Traditions, edt. Hans-Erik Nissen, Heinz K. Klein,
Rudy Hirschheim.
Dretske, F. I. 1981. Knowledge and the Flow of
Information, Oxford, Basil Blackwell
Floridi, L.2005. Is Information Meaningful Data?.
Philosophy and Phenomenological Research, 70.2,
351-370.
Flynn, D. 1998. Information Systems Requirements:
Determination & Analysis 2nd edn. London, McGraw-
Hill.
Gadamer, H. 1975. Truth and Method, Translated by
Barden, G. and Cumming, J.Seabury Press, New York.
Gadamer, H. 1976. Philosophical Hermeneutics,
University of California Press, Berkeley, CA.
Heidegger, M. 1962. Being and Time, Basil Blackwell,
Oxford.
Hirschheim, R., Klein, H.K., and Lyytinen, K.1995.
Information systems development and data modeling,
Cambridge University Press.
Loucopoulos, P., and Zicari, R. 1992. Conceptual
Modeling, Databases, and CASE - An Integrated View
of Information Systems Development. New York, John
Wiley & Sons, Inc.
Mingers, J. 1992. A review of the Soft Systems
Methodology and Information Systems Seminar,
Warwick University, 25th March 1992. Systemist, 14,
3, 79-81.
Mingers, J. 1992. SSM and information systems: an
overview. Systemist, 14, 3, 82-88.
Mingers, J. 1995. Information and meaning: foundations
for an intersubjective account, Info Systems J, 5, 285-
306.
Zhang Ru-Lun.1985. The inquire of meaning:
Figure 3: Reflection process.
Figure 4: The communication pattern between subjects.
THE NEED OF ‘INFORMATION ANALYSIS? FOR INFORMATION SYSTEMS AND OUTLINE OF A
HERMENEUTIC APPROACH TO IT
511