Meaning

Semantics is the study of meaning; the meaning behind the words we use. Semantics enable explainability and automation at scale. Semantics is a core enabler of communication.

In his book, Inhabiting Babel, the author Nicholas Figay makes the following statement in an author's note on page 5: (emphasis is mine)

"Drawing from work in systems engineering, modeling, interoperability, and ontologies, this book seeks to show that meaning is always mediated, situated, and carried by human collectives. Models, graphs, and algorithms do not eliminate this mediation; they displace it and sometimes render it invisible."

* * *

So what precisely does that phrase mean? "Meaning is always mediated, situated, and carried by human collectives."  That sentence is making three tightly linked claims about how meaning actually works in the real world. Breaking that one phrase down into those three claims, for clarity, you get:

“Meaning is mediated”

Meaning never arrives raw. Meaning is always filtered through language, symbols, tools, processes, institutions, and practices.

• You don’t access “reality” directly.
• You access reality through concepts, categories, metaphors, measurement systems, narratives, and shared vocabularies.

So: Meaning is always shaped by the mediums and structures we use to express that meaning.

“Meaning is situated”

Meaning depends on context: social, historical, cultural, professional, technological.

• A word, gesture, or concept means different things in different communities.
• “Arm,” “bank,” “bat,” “park,” “right” each shifts meaning depending on the domain.

So: Meaning is never universal; it’s always anchored in a particular situation, some specific context.

“Meaning is carried by human collectives”

No individual creates meaning alone. Meaning emerges from shared practices, norms, agreements, and interpretations.

• Accounting standards are a collective.
• Scientific concepts are a collective.
• Everyday language is a collective.

People tend to often underestimate this: Meaning lives in communities of stakeholders; not in isolated minds of individuals. And so meaning is more about, it seems, the "shared cognitive understanding and mutual knowledge that people have when they communicate information and interact with each other within a collective,  a.k.a. area of knowledge (i.e. intersubjectivity).

Putting it all together, the initial sentence is saying:

Meaning is not something inherent in just words or objects. Meaning is produced through shared human systems, shaped by context, and sustained by communities of stakeholders.

This appears to be similar to the message of David Weinberger in his book Everything Is Miscellaneous.

Meaning seems to be the bridge between information and knowledge. Data does not seem to have meaning, it has the potential to have meaning. Meaning seems to become into being at the level of knowledge.  A signal is the physical carrier of communication. Breaking this down:

• Noise is the random, irrelevant, or misleading variation that obscures the signal.
• Signal is the medium which has a measurable pattern which becomes data when the data is captured, encoded, and given symbolic form.
• Data is raw and unprocessed and tends to be understandable only in one, usually local, context. Data is encoded and captured signal; but an unorganized and uncontextualized signal. Data includes things like measurements, observations, statements, symbols, and messages.
• Available information is data that has been received, processed, classified, organized, and put into context, therefore potentially interpretable globally by anyone or machine. Information is an organized, structured, and contextualized data. Available information is not yet believed, not yet integrated, and not yet acted upon. It is simply available content. Available information can be true, false, incomplete, or ambiguous. It is abundant and cheap. It has no inherent authority.
• Trusted information: Not all available information advances to become knowledge. To advance to become knowledge, information must be trusted.  Trust is the confidence in the reliability of that information. Trust is the confidence that information is accurate, information is valid per some reliable source or authority, and the interpretation of the information is accepted, shared, and therefore reliable within the conceptual framework. Trusted information only becomes information when it fits into a coherent model, is consistent over time, it is supported by evidence, and it predicts outcomes reliably. Trust is both a lens and a filter because it determines which information we seek, which information we rely on, which information we ignore, and how ambiguous information is interpreted.
• Knowledge is all accumulated trusted information which has been interpreted, internalized, justified, and understood within the scope of some domain of understanding (a.k.a. community, shared definition; shared understanding) giving the information meaning. Knowledge is trusted information that has been refined and that has been further processed, further organized, interpreted per some prior understanding and/or intent, and/or structured in some way making it valuable and actionable. Knowledge is interpreted information that is understood within some specific domain. Knowledge is information + interpretation + justification. Knowledge is scarcer than information, more stable, actionable, but context dependent.  Knowledge is what a community understands; understanding which is justifiably trustworthy. (Know how, know that)
• Insight and wisdom is even more valuable and come from applying knowledge to some specific situation such as making a decision. Understanding is the ability to grasp and work with meaning in a stable, structured way. Understanding organizes meaning, insight transforms or reconfigures understanding and can reveal something which was previously invisible, and wisdom governs or regulates the use of insight.

Information becomes "meaningful" (e.g. information becomes filled with meaning) through interpretation, and that meaningful information becomes knowledge when it is trusted, justified, shared, and made actionable.

Knowledge includes facts, context, contextual metadata.

Information is abundant. Trust is a scare resource.

Knowledge is limited. Modern systems fail not because of lack of information, but because: knowledge becomes fragmented, trust erodes, shared meaning is unstable. The bottleneck in any organizational system is trust, not data.

Information, trust, and knowledge creates a feedback loop or virtuous cycle or a causal chain: trusted knowledge shapes what information we consider valid, which in turn shapes future knowledge.

A domain of understanding or conceptual framework (a.k.a. conceptualization) is the "human collective" doing the "mediation" and the "situating" of information to give that information meaning, agreeing to that meaning per that conceptual framework. Epistemic commitment is the adoption of the conceptual framework. Epistemic risk is the chance that the conceptual framework is wrong.

Communication is the successful alignment of meaning across minds. Communication = Signal + Meaning + Shared Context. Communication can be human to human, human to machine, machine to machine, machine to human.

Meaning is the "interpretable function" that makes information actionable; knowledge is the justifiable, actionable result of the meaning creation process. Meaning arises when information is interpreted within a shared conceptual frame, the "context" or the process of being "situated". Communication is the effective exchange of meaning.

Machine-readable knowledge representations matter because they stabilize meaning so that humans and machines can both reach the same conclusions from the same information. This allows human-machine teams to be created. 

An ontology is a formal, carefully vetted, deliberately and rigorously created model of meaning that is interpretable to both humans and machines.  An ontology describes and explains the things that exist, how those things relate to one another, the properties of those things and relations, and the rules that govern the things and relations. That formal model needs to be complete to properly manage epistemic risk. (Note that I use the term theory rather than ontology to describe the formal model. I see a theory being more complete than an ontology.)

Just because a machine can read something, that does not mean the machine understands anything or that a machine can interpret what it is reading and perform useful work. 

Machines interpret information based on predefined rules.  Humans understand information based on context, skills, experience, research, observation, and reasoning. You cannot automate understanding. 

Understanding is when something makes sense to you well enough that you can use it, explain it, or see how it fits into the context with or connects to other things you know.

Meaning is the content.  Understanding is the competence (skill and experience).

Meaning is static. Understanding is dynamic.

Meaning is an encoding; Understanding is an active process: interpreting, integrating, contextualizing, applying, reasoning.

Meaning is the map. Understanding is the ability to navigate the terrain.

Meaning is intersubjective. Understanding is individual.

Meaning enables understanding. Understanding validates meaning. There is a feedback loop between meaning and understanding.  Without stable meaning, understanding collapses into ambiguity. Without understanding, meaning is inert, no uptake.

Stakeholders in an area of knowledge, such as accounting, have or try to achieve a shared cognitive understanding and mutual knowledge that enables them to communicate information and interact with each other and perform "work". Epistemic risk is the risk that they think they have that shared cognitive understanding and mutual knowledge but in reality they really are not communicating effectively.

Meaning is not inherent. Meaning is an agreement. Meaning exists because of alignment. Meaning is engineered through mediation, context, and collectives. For commerce and the capital markets to function effectively and efficiently for a society, that shared cognitive understanding and mutual knowledge need to be achieved. Meaning is a property of the knowers of knowledge.

A conceptual failure is a misalignment of meaning. Falsification is the intentional manipulation of information.

Intelligence is the ability to learn things, make sense of them, and use those things to solve problems.

Skills determine the resolution and stability of your understanding. Experience is the engine that produces the conditions for insight.

Additional Information:

• Intersubjectivity
• Essence of Accounting
• Conceptualization
• Work
• Carving Up Reality
• Digital Proficiency
• Understanding vs Interpretation
• Intelligence