Printer Friendly

Public and private codes in living systems theory.

INTRODUCTION

The advent of the Internet has brought new salience to the issues of system security and coding. Internet users who will give their credit card number to a stranger in a store may be reluctant to send it across the Internet. One solution is to encrypt or code financial data to prevent it from being stolen on the Internet. Issues of coding and encryption have long been the province of national security agents, remaining remote to the interest of the lay public. The phenomenal growth of the Internet, along with increasing international terrorist threats, has heightened interest in security and coding issues. This has prompted a host of new developments such as the Internet Security Alliance (2003).

The analysis of internal system coding is not just an issue for Internet users, but is paramount to the operation of all systems. Unfortunately, the analysis of such internal coding has been neglected, even within general systems theory. In reality, virtually all systems, at all hierarchical levels, recode both incoming and outgoing information, although the degree and nature of the recoding can vary widely from one system to another. Fortunately, Miller (1978) has analyzed coding within systems in some detail, although his contribution remains unheralded, if not generally neglected.

Internet security can be used to illustrate the role of coding in systems theory. Public key infrastructure (PKI) technology allows a means of securely sending confidential information through the Internet to external destinations (University of Toronto, 2003). The PKI technology relies upon a public key and a private key, similar to Miller's (1978) notions of public and private codes.

The purpose of this paper is to analyze the role of the subsystems of Miller's (1978) living systems theory (LST) in processing public and private codes. The paper distinguishes nine primary data transformation subsystems, which have an active role in the maintenance of public and private codes. The role of coding in providing system security and system identity is emphasized. The relevance of public and private codes to Luhmann's (1989) theory of social autopoiesis is discussed as well. Since the determination of whether a code is called a public code or a private code depends upon the nacre of the system boundary, issues of how boundaries are constructed and maintained, as well as the degree of permeability of a boundary, are also discussed.

In summary, public codes, private codes, and recoding procedures are not limited to the Internet, and must not be left solely to cryptographers. They are generic issues that are crucial for living systems. As systems become more numerous and complex, and more subject to instantaneous computerized interchanges of information, the role of coding becomes even more important. Yet as important as it is, coding is perhaps one of the most neglected aspects of LST. This paper seeks to reduce this neglect through an explication of public codes, private codes, and the recoding process in general.

THE TWENTY SUBSYSTEMS

Originally, living systems were said to comprise 19 critical subsystems (Miller, 1978). Two subsystems process both matter-energy and information. These are the reproducer and the boundary. Eight additional subsystems are strictly matter-energy-processing subsystems. These are: the ingestor, the distributor, the converter, the producer, the matter-energy storage subsystem, the extruder, the motor, and the supporter. The remaining nine original subsystems process information only. These are the input transducer, the internal transducer, the channel and net, the decoder, the associator, the memory, the decider, the encoder, and the output transducer. A tenth information-processing subsystem, the timer, was added later (Miller and Miller, 1992), making a total of 20 subsystems.

Miller applied his analysis of the 20 critical subsystems across eight levels of analysis: the cell, organ, organism, group, organization, community, society, and supranational system (Miller, 1978; Miller and Miller, 1992). Generally, my analysis of public and private codes, and of the recoding process, is meant to apply to all eight levels. However, as space considerations preclude a full analysis of all eight levels here, I will confine many of my examples to the human system. My examples are primarily sociological examples taken from the group, organization, and society levels. At these levels the notions of group security and group identity are particularly salient, as is the notion of social autopoiesis.

Nine of Miller's subsystems are primary data transformation subsystems. This means that not only do they have a central role in transporting or storing the system elements, but also that they actually transform or process the data in some way. These nine primary subsystems include one dual subsystem (the boundary), one matter-energy-processing subsystem (the converter), and seven information-processing subsystems (the associator, the decider, the input transducer, the internal transducer, the output transducer, the decoder, and the encoder).

It could be argued that in some form all 20 subsystems contribute to the tasks of recoding or otherwise transforming the information crucial to system operation. However, for the reproducer, and for seven of the matter-energy processing subsystems (specifically, the ingestor, distributor, producer, storage, extruder, motor, and supporter), their contributions are chiefly secondary. These subsystems focus primarily on supporting materials, or moving them around. Of the nine remaining information-processing subsystems, three are secondary or support subsystems, meaning that they do not directly recode or transform the data in any way. These are the channel and net, memory, and timer. The 11 subsystems just discussed can be termed the secondary support subsystems, as they do not directly contribute to the recoding or transformation of system elements.

THE PRIMARY DATA TRANSFORMATION SUBSYSTEMS

As noted above, the nine primary subsystems are the boundary, converter, decider, input transducer, internal transducer, output transducer, decoder, encoder, and associator. The first primary subsystem that transforms information or data is the boundary. The boundary is 'the subsystem at the perimeter of a system that holds together the components which make up the system, protects them from environmental stresses, and excludes or permits entry to various sorts of matter-energy and information' (Miller, 1978, p. 3). The boundary defines the system by distinguishing it from its environment. The boundary determines whether one is a system member (is inside the boundary), or nonmember (is outside the boundary).

The boundary screens out inappropriate inputs, and allows only appropriate inputs into the system. The inputs that are allowed are those that can foster system goals, while inputs detrimental to the system are rejected at the boundary (Miller, 1978). The boundary is of primary importance in forming and maintaining both the identity and the security of the system. Examples of the boundary for the organization level, as provided by Miller (1978, p. 606), include admissions committees, loan committees, guards, and police.

The second primary data transformation subsystem is the converter. The converter is 'the subsystem which changes certain inputs to the system into forms more useful for the special processes of that particular system' (Miller, 1978, p. 3). Although dealing with materials or energy rather than information, the converter validates the fact that the system is different from the nonsystem (environment), and has special needs. It demonstrates that materials or energy must be transformed from the 'public' form (suitable for transportation over long distances) that they were in before they were input into the system. Examples of the converter for the organization level include canneries, steel mills, and chemical plants (Miller, 1978, p. 607).

The third primary data transformation subsystem is the decider. This is 'the executive subsystem which receives information inputs from all other subsystems and transmits to them information outputs that control the entire system' (Miller, 1978, p. 3). The decider adds instructions about where the information is to go and what is to be done with it. Examples of the decider at the organization level include judges, bishops, and executives (Miller, 1978, p. 607).

The fourth, fifth, and sixth data transformation subsystems are the three transducers. The various transducers all utilize markers. Markers can consist of either matter or energy. They are entities which bear or convey informational symbols (Miller, 1978, p. 12).

The fourth primary data transformation subsystem is the input transducer. This is 'the sensory subsystem which brings markers bearing information into the system, changing them to other matter-energy forms suitable for transmission within it' (Miller, 1978, p. 3). Examples of the input transducer for the organization level include meteorologists, telephone operators, and astronomers (Miller, 1978, p. 606).

The fifth primary data-processing subsystem is the internal transducer. The internal transducer is 'the sensory subsystem which receives, from subsystems or components within the system, markers bearing information about significant alterations in those subsystems or components, changing them to other matter-energy forms of a sort which can be transmitted within it' (Miller, 1978, p. 3). Examples of the internal transducer at the organization level include bookkeepers, payroll departments, and machines such as computers (Miller, 1978, p. 606).

The sixth primary data-processing subsystem is the output transducer. The output transducer is 'the subsystem which puts out markers bearing information from the system, changing markers within the system into other matter-energy forms which can be transmitted over channels in the system's environment' (Miller, 1978, p. 3). Examples of the output transducer at the organization level include public relations personnel and salespeople (Miller, 1978, p. 607).

Each transducer: (a) transmits information; and (b) when necessary, transforms the type of marker that bears the information. The most common types of contemporary markers are paper (book pages), film, pulses of light in fiberglass cable, magnetic pulses transduced from computer floppy disks, electronic impulses used by computers, etc. The importance of the transducer lies not in the fact that it encodes or decodes a message, but rather that it changes the marker by transferring the information from one kind of marker to another. For example, if the employee that is working in the input transducer section of an organization receives a printed document, and decides that no transformation of the information is necessary, he or she simply leaves it in its 'hard copy' form, printed on paper. If transformation to computer form is necessary, modern technology allows the employee several ways of making the transduction. One is to scan the hard copy, another is to retype it on a computer keyboard, and a third is to read it aloud, thus transducing it into computer storage via 'voice recognition' technology.

The seventh primary data-processing subsystem is the decoder. This is 'the subsystem which alters the code of information input to it through the input transducer or internal transducer into a "private" code that can be used internally by the system' (Miller, 1978, p. 3). Examples of the decoder at the organization level include language translators and cryptographers (Miller, 1978, p. 607).

The eighth primary data-processing subsystem is the encoder. The encoder is 'the subsystem which alters the code of information input to it from other information processing subsystems, from a "'private" code used internally by the system into a "public" code which can be interpreted by other systems in its environment' (Miller, 1978, p. 3). Examples of the encoder at the organization level include language translators and lawyers (Miller, 1978, p. 607).

The ninth and last primary data-processing subsystem is the associator. The associator is 'the subsystem which carries out the first stage of the learning process, forming enduring associations among items of information in the system' (Miller, 1978, p. 3). The associator adds associations and interpretations to the message, thereby personalizing it. Miller further describes the associator by saying that 'Consequently it constitutes a private organization of knowledge or wisdom--as opposed to raw items of information--organized especially for the internal needs of that particular system' (Miller, 1978, p. 65). This private organization of knowledge is in a sense a second level of privacy that is structured on top of the first level of privacy (the internal or private code). Examples of the associator at the organization level include outward dispersal to management consultants and operations researchers (Miller, 1978, p. 607).

TYPES OF CODES

Miller (1978) distinguished three main types of codes: alpha codes, beta codes, and gamma codes.

Alpha Codes

The lower-level codes are termed alpha codes. Cells, organs, and the lowest species of animal organisms can only process alpha-coded information (Miller, 1978, p. 404). In cells, alpha-coded information is carried on chemical markers, such as hormones or chemical molecules (Miller, 1978, p. 265). In lower animals, alpha-coded information is carried on molecules called "pheromones' (Miller, 1978, p. 404). These can be transmitted through the air as odors, and processed by other individuals of the same species (Miller, 1978, p. 404).

Beta Codes

While cells, organs, and lower animals are limited to processing alpha-coded information, living systems at higher levels can process beta-coded information (Miller, 1978, p. 265). Beta-coded information is input as patterns of energies such as light, sound waves, or electrical transmissions. Higher animals can all decode alpha codes, as can cells, organs, and organisms.

Gamma Codes

The third type of coding is gamma coding. Miller (1978, p. 404) says that only humans are positively known to have the ability to process gamma-coded information, although some other species such as dolphins and chimpanzees may be able to do so to a limited degree. Gamma coding is the symbolic coding of conceptual or abstract ideas, as conveyed in written language. Miller (1978, p. 635) says 'the "private language" of an organization reflects the categories of meaning or sets of concepts by which information from the environment or from within the organization is classified by its components. Messages in such an internal code of an organization are gamma-coded information' (italics in the original).

At the group level and above in human society, most of the communication involves gamma-coded written or spoken language. However, alpha and beta coding is also used. An example of alpha coding, where messages are conveyed by chemical molecules, includes the use of fragrances such as perfume or incense, or the use of special foods, alcoholic beverages, or drugs in religious ceremonies (Miller, 1978, p. 635). Beta coding is more widely used in organizations than is alpha coding, but not as widely used as gamma coding. Examples of beta-coded messages include a variety of non-linguistic symbolic transmissions such as sign language, 'body language', music, ritualistic behavior, and the wearing of uniforms (Miller, 1978, p. 635).

As stated earlier, one purpose of this paper is to explicate Miller's analysis of public and private coding by comparison with PKI and other features of Internet coding and security. Since the computer codes used in Internet transmission are exclusively highly abstract gamma codes, the focus of this paper will be on gamma coding at the human group, organization, community, society, and supranational levels. Thus, the advantages and disadvantages of recoding discussed below are focused primarily on human systems. Recoding in systems at the cell, organ, and organism levels may have a somewhat different set of advantages and disadvantages.

It should be pointed out that gamma coding does not replace beta and alpha coding, but is used in conjunction with it. Thus, while Internet transmission codes are gamma codes, beta codes frequently appear on the computer screen as icons, pictures, and colors. Further research on alpha and beta coding is clearly needed, but cannot be included in this short paper.

DATA TRANSFORMATION, CODING, AND RECODING

Terms such as encoding, decoding, and encryption are prevalent in the literature. However, these terms can be confusing, and may be used inconsistently in the literature. Thus, it behooves us at this point to briefly clarify or define some basic terms used in this paper. The problem with terms such as encoding or decoding is that they are indexicals (Leiter, 1980). An indexical is a term or phrase that is so context-specific that its meaning is unclear when taken out of context. Common indexicals such as 'he', 'she', and 'it' abound in everyday parlance, but can only be interpreted within a given context. The same is true with terms such as 'encoding' and 'decoding'. Encoding means to take a message that is in 'old code' (its original form), and put it into 'new code'. Decoding means to take the message out of the new code, and put it into old code.

The problem is that every message already is in a 'code' of some sort. Thus, a person that is 'encoding" into a new code is generally simultaneously 'decoding" from the old code. Further, due to the indexical nature of these terms, one generally must know their context in order to determine whether the process should be called encoding or decoding. Encryption and decryption are a little different, as these terms are often limited to the process of recoding from an alphabetical code into a numerical (generally a 'scrambled') code, and they can probably be retained with some clarity, although strictly speaking they too are indexicals.

In addition to recoding, the term 'coding' is often used. Strictly speaking, coding would only apply to a situation when there previously was no code at all. This would only apply to the writing of a new message, not to a message that is being received, as all of these messages were in a code of some sort.

An even more inclusive term is that of data transformation or information transformation. This is a very general term that can be applied to any case of message alteration, whether or not there is any change of meaning, or loss or gain of information. Decision making is data transformation if it involves altering the message, or shortening or lengthening it. Further, recoding, transducing, associating, and converting all transform information in some way, so all of the nine subsystems discussed above are involved in data transformation.

The very practice of referring to a recoding subsystem as either a decoder or encoder remains misleading, as it masks the fact that each subsystem performs both decoding and encoding functions when processing a single message. The decoder decodes public information coming into the system from outside, and encodes it into private information for internal use by the system. The encoder reverses the process. It decodes the internal private information, and encodes it into public information that can be sent outside the system.

I will largely eschew the terms decoding and encoding in this next section where I discuss the advantages and disadvantages of recoding, but will employ the more generic and comprehensive term of recoding instead. As I use the term, recoding means changing from one code to another. This applies to both encoding and decoding.

ADVANTAGES AND DISADVANTAGES OF RECODING

The decoding and encoding functions can occur quite rapidly, one after the other, and can seem meaningless or redundant to some observers. Why should I take a message from an external system, and bother to 'decode' it for internal private usage, only to subsequently be required to "encode" it once again, back into roughly its original state, for retransmission back to the external system? How can I justify the expenditure of time and money that recoding entails? Why even bother to recode at all? Could we not simply dispense with the whole decoding/ encoding process, and retain the information as it was? The answer is, absolutely not. This is one of the least understood features of systems.

The key to understanding this aspect of systems is to understand that if coding were unnecessary, then the entropy of the system would be equal to the entropy of its environment. In this case, the boundary would not represent an entropy break (Bailey, 1990, 1994), but would be merely a signpost, with little real significance. To say it another way, if the system did not use private code, but only public code, it could easily become indistinguishable from its external environment. Even if recoding into private jargon were unnecessary for communication purposes, it still must be done to distinguish the system from its environment.

But while recoding is a necessary task for systems, this does not mean that there are no disadvantages to it. Focusing on human social systems that are using primarily gamma coding, it is seen that there are at least seven advantages and four disadvantages of recoding.

Advantages of Recoding

(1) Security. A major reason for recoding is to protect internal information from external entities. This reason is so important that it is deserving of its own section (along with identity), and so is discussed in detail below.

(2) Identity. Another very important reason to recode, and one that is highly correlated with security, is the formation and maintenance of internal system identity. This important function of recoding is for some reason often neglected in the literature, and so is also discussed in detail below.

(3) Internal uniformity and standardization. Some systems may need to ensure uniformity of message format or wording (or both) in order for their internal communication procedures to operate efficiently. In fact, this reason seems to be the operative one that Miller (1978) had in mind when discussing decoding. He was concerned that the message be recoded into a form that could be efficiently processed within the system by the other subsystems, such as the internal transducer and the decider. In this case, recoding messages to meet systems standards and practice is necessary in order for internal information-processing procedures to work properly.

(4) Reduction of external message diversity. Another advantage of internal recoding (and one that is clearly related to that of uniformity, but remains conceptually distinct), is the reduction of external diversity. If a system routinely receives the same sort of message from all of its external sources (even if it has many different sources), or if it has only one or a few external message sources, the task of recoding to achieve internal message uniformity may still be necessary, but may be simple to achieve, or capable of routinization. However, if the system receives communications from a wide variety of diverse external sources, or if it frequently adds new external message sources, then some recoding procedures that seek to reduce diversity of both message content and format may be necessary, even prior to recoding for uniformity, as mentioned above. That is, multiple-stage recoding procedures may be necessary in some cases, with the initial recoding perhaps being predominantly a screening device of some sort.

(5) Autonomy and control. Even if there is little diversity of format or content in incoming messages from external sources, and even if the incoming messages could be adequately processed without recoding, some systems may wish to recode into their own special words, symbols, or message format simply to retain autonomy and control, and to differentiate themselves from outside sources. Such control may aid in the reduction of message diversity and the subsequent establishment of uniform standards for message processing, as discussed above. It may also save resources by making message processing more efficient, and by reducing processing errors, or even the reduction of errors already imbedded in messages received from external sources. It may also promote a general feeling of self-reliance and well-being, and thus contribute to internal identity, to be discussed below.

(6) Comprehension. Ultimately, the main reason for recoding is to increase comprehension, and to facilitate overall system functioning, including the functioning of all other relevant subsystems, as discussed above. This includes comprehension not only on the part of the internal receiving system (in the case of incoming messages), but also comprehension by external receivers (in the case of outgoing messages).

(7) Overcoming system limitations. Yet another reason to recode is to cope with the limitations of the system. This is particularly true for lower-level systems, such as those at the cell and organ level. Systems at these lower levels are not as complex as at higher levels, and do not have the ability to process gamma codes. These lower-level systems have many limitations and must recode the chemical information that is input to them in order to cope. They are not recoding from one level of code to another (e.g., from the alpha level to the gamma level), but are merely recoding information that is input to them at the alpha level (or sometimes the beta level). For example, the cell may receive information from the environment on chemical markers such as hormones (Miller, 1978, p. 265). This information then 'may be recoded one or more times inside the cell' (Miller, 1978, p. 265). Note that recoding is not the only way to overcome system limitations. Such limitations may also be overcome by using different kinds of markers, or by code simplification.

Disadvantages of Recoding

(1) Time. One of the major disadvantages of recoding is the time spent in this process. This can vary widely, depending upon the complexity of the recoding process for the particular message. In cases where the time spent on recoding seems excessive, it is understandable that some managers may wish to streamline this step.

(2) Expense. Another major disadvantage of recoding within the system is the inevitable financial or energy cost. This may be substantial if it necessitates the hiring of experts for language translation or mathematical encryption, for example.

(3) The inadvertent introduction of error. Recoding may induce error if done improperly. Incompetent, hasty, or inconsistent recoding may result in the introduction of errors, both substantive and clerical. However, errors may be more likely to occur in the case of complex recoding tasks such as language translation or mathematical decryption. In such complex cases, the introduction of at least some minimal amount of error in recoding is perhaps inevitable, and must be tolerated.

(4) Alienation of allies. In some cases, external allies may object to the recoding of their messages. They may feel that the recoding agency is treating their message as flawed, and thus in need of revamping to improve clarity. Alternatively, the senders may feel that their message is being recoded in order to make its analysis secret, and thus no longer available to them. This may be alienating, making them feel like an outsider, rather than like a full partner in the communication process.

SOCIOLOGICAL EXAMPLES

At this point it would be preferable to illustrate the coding process through explications of alpha, beta and gamma coding, at all levels from the cell to the supranational system. Since space clearly precludes this, I must choose some more limited illustrations. In keeping with my discussion of the parallels between Internet security and Miller's public and private codes, it seems wise to concentrate on the social system. I will deal primarily with system security and system identity, with some discussion of the related topic of information transmission across boundaries.

Internal System Security

One chief reason for having decoders, encoders (and for having boundaries, converters, and transducers also) is to maintain system security. This is clearest for financial information, but also for trade secrets, patent data, intellectual property, etc. If this information is made public, then the whole future existence of the system may be impaired. The dilemma is that while internal security must be maintained (through the use of the private code), the system must also simultaneously retain its ability to communicate with external systems in the environment, and this necessitates retention of the public code. These two needs--the need for private security and public communication--are essentially contradictory, and thus necessitate the subsystem processes of boundary, converter, transducer, associator, and recoder.

The need for decoding and encoding has been dramatized by the advent of the Internet. While you may feel perfectly capable of protecting the security of your credit card number in your internal home system, you can only buy goods online by sending your credit card number externally, into the vastness of the Internet environment. Many people are unwilling to do this, and so purchasing on the Internet has been slowed.

The answer is a mirror of Miller's (1978) private and public codes--public key infrastructure (PKI). PKI works by giving the individual system a private key, and also establishing a public key (University of Toronto, 2003). Your credit card number is sent to the public key, where the merchant can safely decode it. However, this is only done after the number is protected by coding it through your private key. This process of decoding private information would generally only be done inside the system. By letting the external merchant decode your credit card information, you are in effect granting him or her trusted 'insider' status, if only temporarily. The coding process depends on mathematical cryptography, or 'scrambling" of the credit card data. The more complex the cryptography, the safer is the coded information. For example, while 4-bit (binary digit) encryption would be relatively easy to decode, higher-order encryption becomes very difficult to decode.

Currently, the exact degree and manner of encryption can vary widely with the particular system, but 128-bit encryption ([2.sup.7]) is commonly used, and considered to be generally adequate. One could use 256-bit, 512-bit, or even 1024-bit encryption ([2.sup.10]), or more, but this would be unnecessarily cumbersome. The typical PKI system, if properly constructed, is extremely secure, although the public is still wary of it, as it is very technical and is not well understood. It works because of the decoding/encoding of public/private codes, as illustrated in Miller's (1978) subsystems. When your computer screen shows a picture of a lock, and displays an 's' at the end of http [https] in the Web address, then you know that an Internet security system is in place.

Internal System Identity

In addition to security issues, identity is also crucial to the system. A central function of the living system is to distinguish system members from non-members; that is, to distinguish 'insiders' from 'outsiders', 'they' from 'us', etc. The chief way that this is accomplished is by distinguishing between a shared public external code that 'they' (the 'outsiders') can possess and understand, and the private code that only 'we' (the system 'insiders') can possess or understand.

If the system had no private code, it would rely solely on public code. In this case, its entropy level would be equal to (or as high as) the external environmental entropy level and the internal system would have no unique identity. One could then not easily discern whether he or she was in the system or in the external environment, as there would be no real difference. Thus, the system must recode external public code into a private code to establish system uniqueness, even if this seems unnecessary to outsiders. If I am a system insider, I rely on the private system internal code, not only to ensure privacy from outsiders, but also to distinguish insiders from outsiders. If you speak the insider private code language, I know you are an insider, and this gives me a sense of affinity with you. If you do not speak the insider private code, but only use generic public code, I can identify you as an outsider.

There are two chief ways that problems with private code can lead to system identity crises. One way is if private code is allowed to deteriorate, or if members abandon it in favor of generic public code. This clearly threatens group identity. This explains why ethnic groups often seek language purity, and hate for words from foreign languages to enter their language. Another threat is when outsiders co-opt private code, and attempt to incorporate it into public code. This often happens because outsiders yearn for acceptance. They seek group membership or insider status. They wish to appear 'hip' by adopting insider code from insider groups such as teenage cliques, criminal gangs, or ethnic minorities. As soon as the interlopers secure knowledge of the private code and begin using it, it becomes tainted. It loses much of its value as an indicator of insider status or identity, and is generally abandoned by the insider group.

One prime example of age-graded teenage private code is insider knowledge of the names of musical groups that are currently popular with young people. As soon as older adults learn the names and lore of these groups, the groups can quickly lose favor, and be subsequently abandoned by the insider teenagers, as private code has now entered the public code. System entropy has risen to unacceptable levels, making it more difficult to differentiate between insiders and outsiders on the basis of code usage.

All systems face a dilemma concerning what to do when their private code is corrupted, or co-opted by outsiders. If the system is a relatively transitional or ephemeral one like a teenage clique, the easiest solution is for the insiders to simply abandon the tainted private code, and replace it with fresh private code. But the situation is more grave if the threatened private code represents core concepts in a discipline. This is in a sense the dilemma confronted by early systems theory. The founders of systems theory reasoned (correctly) that there was redundancy among disciplines that could be reduced by integrating them into a single systems theory.

However, problems arose when attempts by general systems theory (GST) to co-opt concepts from various disciplines were seen by disciplinary insiders as a threat to private code. In other words, while systems theorists may view a particular term as part of the public code (or as part of the larger, systems theory private code that is not unique to a specific discipline), disciplinary insiders may view that specific term as part of their discipline's unique, proprietary private code.

Consider the example of entropy. When the term was invented by Clausius in the nineteenth century, it was coined from an ancient Greek word (see Coming and Kline, 1998, p. 274). This relatively public (Greek) code was thus converted into private thermodynamics code. Recently, entropy has been widely used outside of thermodynamics, particularly in systems theory. Yet, many students of thermodynamics still resist exportation of such a central theoretical term to the external public code (Bailey, 2001). They are in effect acting as though entropy is still private code, and thus should not be used outside of thermodynamics.

Systems theorists could reply that while using entropy in an admittedly more generic sense, they are still using it as 'private code' (albeit an expanded private code) within the expanded private sphere of systems theory. However, this view is being increasingly weakened by the use of entropy in the public code on the Internet, on license plates, lapel buttons, or in business names. There is no doubt that the frequent public use of the term 'entropy' has led to some misuse and degradation of the concept (Bailey, 2001). This is evidenced by the frequent misspelling of the term as 'entrophy' (perhaps through confusion with 'atrophy').

This constitutes a clear dilemma for thermodynamics insiders who still view entropy not only as a private code term but as a core concept. If they abandon entropy or replace the term, they will risk damaging a large and venerable body of thermodynamic theory. On the other hand, if entropy is allowed to become a public code term, this not only potentially weakens the integrity of the original concept, but also poses the threat of theoretical deterioration for thermodynamics. While no full solution to this dilemma is currently at hand, at least an interim solution is to use adjectives that recognize different subtypes or applications of entropy. Thus, 'thermodynamic entropy' ('Clausius" entropy,) can retain its standing as a private code in thermodynamics, without being confused with other types of entropy such as 'Shannon's entropy' (Shannon and Weaver, 1949), or 'social entropy' (Bailey, 1990, 1994).

AUTOPOIESIS

Autopoiesis is a word coined by Maturana and Varela (1980) to mean 'self-production'. Marurana and Varela established that autopoiesis exists on the level of the cell, as cells clearly contain the processes needed to self-produce. However, a controversy remains as to whether human systems at the level of the group and above are autopoietic or not (see Luhmann, 1989, 1995; Mingers, 1995; Robb, 1989).

The notion of autopoiesis assumes that the system has an internal organization that must be produced and preserved. This is accomplished through reliance on private codes. Luhmann (1989, 1995) demonstrated that all professions (e.g., medicine, law, and science) rely heavily on private codes. These codes take the form of unique professional jargon. This private code jargon is used not only to further the autopoiesis of the profession, but also to keep outsiders at bay.

Only accredited insider professionals with full credentials are licensed to use the private legal code or medical code. Outsiders who use such insider code in an unauthorized manner may be formally charged with practicing medicine without a license, practicing law without a license, etc. Outsiders may communicate with the professional insiders, but generally only do so via public code. The insider then recodes public code into private code (e.g., medical transcription), and uses the insider code to fulfill the external client's request. After the task is completed through use of the private code, the professional then recodes (encodes) back into public code to respond to the outsider.

As an example, the medical patient describes symptoms to the physician in public code. The physician then makes a diagnosis in private code (decoder), and records the diagnosis in the patient's medical chart, (input transducer), which can then be transmitted to other medical professionals (internal transducer). The information is ultimately encoded for use by the patient or others outside of the medical profession, and is transmitted from the medical system to the outside environment via the output transducer. Through the processes of transducing, decoding, and encoding, the autopoiesis of the medical profession can be furthered, while simultaneously serving the needs of the external clients (the patients).

CONCLUDING REMARKS

This paper discussed the neglected topics of public and private codes, as well as the related processes carried out by the boundary, decider, converter, decoder, encoder, associator, and transducers. It was shown that the public and private codes play a central role in defining and maintaining systems, and in facilitating communication between the system and its environment. It was shown further that recoding has a number of advantages (as well as some disadvantages) in the maintenance of public and private codes. In particular, the maintenance of separate public and private codes is instrumental not only in maintaining system security but also in maintaining system identity. Further, public and private codes are helpful in understanding autopoiesis.

It should be noted in closing that the recoding of messages in and out of the system, as discussed in this paper, is part of a larger issue in modern systems theory--that of the relationship of a given system to its external environment, including other systems in that environment. For too long we have emphasized system internals, and have neglected the analysis of system externals, and the relationships between them. Luhmann (1989, 1995) has called for increased emphasis on the analysis of the relationship between a system and its external environment (including other systems in that environment), rather than a continued focus on system internals.

The role of public and private codes is extremely important in the larger context of system--environment relations, and serves as a convenient starting point for this larger analysis. Unfortunately the systematic study of these codes has been sorely neglected. It is important that the notions of public and private codes, and the attendant issues of system security and identity (as well as the other recoding issues discussed above), be more carefully studied in the future.

REFERENCES

Bailey KD. 1990. Social Entropy Theory. State University of New York Press: Albany, NY.

Bailey KD. 1994. Sociology and the New Systems Theory: Toward a Theoretical Synthesis. State University of New York Press: Albany, NY.

Bailey KD. 2001. Towards unifying science: applying concepts across disciplinary boundaries. Systems Research and Behavioral Science 18: 41-62.

Corning PA, Kline SJ. 1998. Thermodynamics, information, and life revisited--Part I: 'to be or entropy'. Systems Research and Behavioral Science 15: 273-295.

Internet Security Alliance. 2003. ISA knowledgebase on wireless technologies, http://www.isalliance. org/ [25 October 2003].

Leiter K. 1980. A Primer on Ethnomethodology. Oxford University Press: New York.

Luhmann N. 1989. Ecological Communication. Polity Press: Cambridge, UK.

Luhmann N. 1995. Social Systems. Stanford University Press: Stanford, CA.

Maturana HR, Varela FG. 1980. Autopoiesis and Cognition: The Realization of the Living. Reideh Dordrecht.

Miller JG. 1978. Living Systems. McGraw-Hill: New York.

Miller JL, Miller JG. 1992. Greater than the sum of its parts--I: subsystems which process both matter-energy and information. Behavioral Science 37: 1-38.

Mingers J. 1995. Self-Producing Systems: Implications and Applications of Autopoiesis. Plenum: London.

Robb F. 1989. Cybernetics and supra human autopoietic systems. Systems Practice 2: 349-351.

Shannon CE, Weaver W. 1949. The Mathematical Theory of Communication. University of Illinois Press: Urbana, IL.

University of Toronto. 2003. Securing Access to the Internet: PIG Assessment Project. http://www. utoronto.ca/security/[25 October 2003].

Kenneth D. Bailey *

Department of Sociology, University of California, Los Angeles, California, USA

* Correspondence to: Kenneth D. Bailey, Department of Sociology, 405 Hilgard Avenue, University of California, Los Angeles, CA 90095, USA. E-mail: kbailey@soc.ucla.edu
COPYRIGHT 2006 John Wiley & Sons, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Research Paper
Author:Bailey, Kenneth D.
Publication:Systems Research and Behavioral Science
Geographic Code:1USA
Date:Nov 1, 2006
Words:6814
Previous Article:Editorial.
Next Article:Working at the interface between systems and complexity thinking: insights from a market access design project for poor livestock producers.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters