Reforming the Ogden-Richards Triangle
for Meaning

Oliver Objectivist: Dictionary makers don’t have to go into all the things that philosophers fight about. They just do their job. So in a dictionary we have information that doesn’t depend either on philosophy or on religious conviction, but merely on technical competence.

Herman Hermeneut: But don’t dictionary makers, either tacitly or explicitly, have a certain framework for their understanding of what they are doing? Don’t they make some key assumptions about how language operates, what words are, what meanings are, and so on?

Objectivist: Yes, but they can agree. Years ago Ogden and Richards set out the basic theoretical framework, and it has proved serviceable ever since.

Herman Hermeneut: Let’s look at it.

True worship and idolatry affect the details of interpretation as well as its global structure. As one small example of these effects, we may consider the Ogden-Richards triangle.

The Ogden-Richards triangle

The Ogden-Richards’s triangle offers a model for the nature of linguistic symbols.¹ The original diagram is reproduced below.

The three principal elements are ‘‘symbol,’’ ‘‘thought or reference,’’ and ‘‘referent.’’ The symbol is a particular phonological or graphical form such as e l e p h a n t. The thought or reference is the meaning-content of the form, namely the idea of an elephant. The referent is the nonlinguistic thing in the world being talked about (actual elephants). Stephen Ullmann prefers to relabel the vertices of the triangle as ‘‘name,’’ ‘‘sense,’’ and ‘‘thing,’’ respectively.² Moisés Silva labels the vertices ‘‘symbol,’’ ‘‘sense,’’ and ‘‘referent.’’³

Our own framework can account for this natural division. But for that purpose we need to use several kinds of distinction. First, recall that God’s purposes toward us involve control, meaning, and presence (as previously defined). He controls what happens in the world, that is, he creates and controls referents. He does so through his word, which expresses meaning. The meaning includes both thought and symbol. Thus, we may correlate the referent with God’s control. We correlate thought and symbol with the meaning aspect. So far, from the triad of control, meaning, and presence, we have used two aspects, control and meaning. The Ogden-Richards triangle has left out completely the third indispensable aspect, namely the person of the language-user. A careful description would then reemphasize this excluded aspect.

Now thought and symbol both belong to the meaning aspect. So what is the difference between them? Thought concerns the things about which we think and speak. Symbol concerns the things by means of which we speak, that is, sounds or letters, or perhaps still other symbols such as Morse code or sign language.

Symbol, then, depends on the media within the created world that human beings use in communication. But these media, though created, have a divine archetype. Starting from John 1:1, we affirm that God the Father is the original Speaker, while God the Son is the original Word. The Holy Spirit is the original Breath carrying the Word to his destination (Ps. 33:6; compare Ezek. 37). Within the created sphere, the analogue of the Spirit is the created breath of human utterers. The word goes forth as sound in the air. We therefore claim that the sound aspect, or graphical aspect, is built by analogy on the Spirit. The thought aspect is built by analogy on God the Father, whose thought is then expressed in the Word. The lingual expression as opposed to the thought is analogous to God the Son, the Word. In sum, we have three aspects, thought, lingual expression, and sound, where the Ogden-Richards triangle has only two, thought and sound.

Ullmann and Silva, unlike Ogden and Richards, label the top vertex as “sense” rather than “thought or reference.” Thereby they focus on the lingual expression; they leave the mental activity, the thought, in the background. This alteration of terminology is useful, in that it more clearly accomplishes what Ogden and Richards probably wanted, namely to put all the focus and emphasis on the linguistic aspect. Whichever label we use, within a Christian framework we recognize a coinherent relation among thought, discourse, and sound.

To understand the triangle more precisely, we need one further distinction. Ogden and Richards did not intend that the triangle should merely represent the function of one particular utterance, but rather the systematic function of sense, symbol, and referent in the whole of a particular language. They wanted to say something about the internal structure of a language system. Rather than dealing with one particular utterance, they are concerned with the general pattern of usage summarized in a dictionary or a grammar. The dictionary records that elephant means an animal of the family Elephantidae throughout all its various occurrences. Speakers of English must know about the meaning of the word in general in order to understand it in particular occurrences. That is, they already know a language, including the facts reported in a dictionary.

As we might expect, the particularities of utterances and the generalities of a language are correlative to one another. The particularities show the instantiational aspect, while the generalities show the classificational aspect. Both the particularities and the generalities enjoy a system of relations with all the other particularities and generalities, thus forming a system of relations. The relations manifest the associational aspect. Since the instantiational, classificational, and associative aspect image the Trinity, the distinction among utterances, generalities, and systematic relations also images the Trinity.

It follows, then, that just as the Persons of Trinity coinhere, language generalities, utterances, and systemic relations coinhere. Thought, linguistic expression, and sound realization coinhere. Referents, meanings, and speaking persons coinhere.

Since God is one God, there is unity in language. A linguistic symbol is what it is in a unity through the joint presence of symbol, thought, and referent. Since God is Triune, there exists also the possibility of distinguishing between symbol, thought, and referent.

Hence, the triangle can aid in identifying certain fallacies. The dotted baseline connecting symbol and referent, plus the additional vertex labeled ‘‘thought,’’ remind us that the relation between symbol and referent is indirect. Errors can arise if we ignore this complexity. In a word-and-thing approach, we may naively equate a symbol with its referent. For example, when we deal with abstract theological words like ‘‘righteousness,’’ the referent is similar to what we might call a theological ‘‘concept.’’ Hence we equate word and concept. We then fall into confusions like the ‘‘historico-conceptual method typified by TDNT.’’⁴ For example, we might reason that because the Hebrew word dabar may refer to a dynamic historical event, it must mean ‘‘dynamic historical event.’’ But it actually means ‘‘word, thing, matter,’’ a quite colorless designation.⁵

Hence, the triangle is a useful summary of some distinctions that need to be borne in mind in interpretation. But we need also to recognize that the triangle is a simplification, in several respects.

(1) It does not address the mysteries residing in the relations between communicators and their utterances. The aspect of presence, and with it the participation of speakers, is left out.

(2) It does not explicitly explore the mysteries involved in the relation of thought to language and sense. It can give the illusion that sense and thought could in principle be isolated from one another.

(3) It does distinguish explicitly between particular utterances, linguistic generalities, and relations of these to other linguistic regularities. It thereby runs some danger of confusion. For example, the label ‘‘symbol’’ could suggest focusing on a particular occurrence of ‘‘steal’’ in a particular copy of a particular English version of Ephesians 4:28. It would thus focus on a particular discourse. Or the same label could cover all occurrences of ‘‘steal’’ in all actual or even possible contexts. It would thus focus on the language generalities. Or it could examine the relation of “steal” to other words in the same semantic domain.

(4) The triangle places in the background the contextual dimensions of communication. But in fact, meaning exists only in close connection with the operation of larger context, as we have seen in our reflection on the coinherent relation between “sense” and application and import. The diagram proposes to analyze the function of meaning attaching to a unit, whether a word, a morpheme, or a longer expression. But units are perspectivally related to contexts and hierarchies, producing an import.

(5) The triangle can easily suggest that the symbol and the sense are neatly separable from one another. But remember that symbol focuses on the aspect of utterance, while sense focuses on the aspect of meaning. In any particular case these are two aspects of the same communication. The two poles are not neatly separable. All meaning comes in ‘‘form-meaning composites,’’ as Pike has argued.⁶

Hence, symbol and sense are perspectivally related. Each irreducibly involves the other, by virtue of coinherence. A symbol is not a symbol unless it symbolizes, that is, it already has sense that is inseparable from its identification as a symbol. Conversely, any particular sense is recognizable and identifiable as the same only through media, that is, through symbol.

When Ullmann and Silva try to define the upper vertex of the triangle more closely, the involvement of symbol with sense becomes even more apparent. Ullmann says that sense is ‘‘the information which the name conveys to the hearer.’’⁷ This definition certainly invites us to focus on truth content or thought content. But in a broad sense a symbol ‘‘conveys information’’ about its own phonological form. Such information is not merely trivial, but must be utilized by any language learner or interpreter. Hence, the shape of the symbol as symbol is also part of the information conveyed. Silva defines sense as ‘‘the mental content called up by the symbol,’’ where the words ‘‘mental’’ and ‘‘content’’ once again zero in on speakers and their thoughts. But in fact our ‘‘mental contents’’ in a broad sense include the loose associations between dogs and the sound of the word dog. Thus, the distinctions in this area do not really succeed in distinguishing in some absolutely definitive way between symbol and sense.

(6) Since the referent is ‘‘non-linguistic,’’ it is tempting to exclude the referent from linguistic analysis and to retain only the symbol and the sense.⁸ The exclusion rests on the valid intuition that the situation and the understanding by persons in the situation are perspectivally distinguishable. Ullmann provides an example:

The atom [the referent] is the same as it was fifty years ago, but since it has been split we know that it is not the smallest constituent of matter, as etymology suggests; moreover, it [the sense] has been enriched with new connotations, some fascinating, others terrifying, since the advent of the atomic age—and the atomic bomb.⁹

But again distinguishable aspects are not separable. Meaning depends on persons who communicate, and persons depend on a situation and norms about which to talk. Hence, the referent cannot be eliminated from analysis.

These last two points are especially important. In point (5) we assert that symbol and sense (form and meaning) are perspectivally distinguishable but not separable. In point (6) we assert that sense and referent are perspectivally distinguishable but not separable. Point (6) can be made even more strongly by considering God’s speech, in contrast with human speech. God through his kingly rule governs all the facts, that is, all the referents. All the referents are what they are in response to God’s words of decree. God’s word governs the facts. The facts obey the word. To have facts outside God’s word would be impossible, meaningless. Thus, having meaningful referents for human beings depends on having referents whose every aspect is subjected to God’s word, to his ‘‘sense.’’ All referents are in an ultimate analysis completely comprehended by sense. Of course, the finite understanding of sense by human beings does not exhaustively grasp referents. But language as a vehicle for divine communication is not exhausted by the human understanding of it.

The roots of linguistic distinctions

Another major limitation of the Ogden-Richards diagram or of any analogous diagram is that it cannot achieve ultimate explanation of its distinctions. For example, the diagram expresses the fact that symbol and sense are distinguishable as form and content. Well and good. But how do we so distinguish them? Semanticists can bring forward any number of illustrations to help us grasp the distinction. They say, “The spelling of dog as d-o-g is the symbol, and the sense is the marking out of an area of meaning, such as ‘an animal of the species Canis familiaris.’ ” But what makes it possible to see the general pattern from the illustrations? We intuitively know that form and content are distinct, because we already have the capacity to distinguish them. Illustrations and explanations always presuppose prior capacity to ‘‘see the point.’’ Ultimately we have this capacity because of some unanalyzed commonness in human nature. But human nature itself does not exist in a vacuum. We know what content is because we know God’s truthfulness, that is, the thought of the Father and the expression in the Son. We know what linguistic form is because we know God’s articulate discourse through the Spirit.

Similarly, the distinction between sense and referent is valid, but what are its roots? This distinction in human communication images the distinction in divine communication between God’s decree (sense) and the created things obeying his decree (referent).

These two distinctions thus ultimately go back to distinctions about God and the nature of his interaction with the world. The ways in which these two distinctions refer back to God is typical. We cannot understand a single distinction introduced by semanticists without the personal presence of God in his Triunity.

There is still an additional complexity. Since the Persons of the Trinity coinhere, the three vertices of the Ogden-Richards triangle coinhere by analogy. Most semanticists have not really admitted this fact to themselves, but continue to act to some extent as if ideally the vertices were perfectly separable. We may suspect that idolatry is a factor in this distortion. In unitarian idolatry, to have knowledge we must have knowledge of unities with no remaining diversity; that is, we must have isolatable unities.¹⁰ The isolation of unities then also reflects a kind of polytheism, because there are still many unities. Because we live in God’s world and continue to know God in spite of ourselves, these implications never become as destructive of knowledge as they could be. But counterfeiting still produces mixtures of truth and error throughout the analysis of language. Counterfeiting distorts our knowledge of the Trinitarian God. This distortion affects our knowledge of the Ogden-Richards triangle, and this distortion in turn affects the knowledge of the sense of every word in our language. The effects are subtle but pervasive.

Thus, the differences in our knowledge of God affect all aspects of interpretation, including even the more technical aspects such as the study of the sense of a particular word. Christians and non-Christians differ radically in their knowledge of God (Eph. 4:17-24). Hence they differ radically and inevitably in interpretation. The differences affect biblical interpretation first of all and most obviously. But the same arguments lead to the conclusion that differences in knowing God affect interpretation of any document or any created thing.