A Survey on the Relational Processes of experiential metafunction

This processes indicate the relationship between one entity and another， or refer to one entitys characteristics or property.There are two types of relational processes：attributive and identifying. Each has two participants， namely Carrier and Attribute and Identified and Identifier.

As this is a rich and complex area of clause transitivity， we only make an outline here， and begin by clarifying the basic structural difference between Attributive and Identifying process.

1. Attributive Relational Processes

The relational process involves establishing a relationship between two terms， where the relationship is expressed by the verb “be” or a synonym. In the Attributive sub-type， a quality， classification or descriptive epithet （Attribute） is assigned to a participant （carrier）. The Carrier is always realized by a noun or nominal group.

In the classification kind of Attributive relational process， the Attribute is typically an indefinite nominal group. The meaning of this kind of relational process is that “X is a member of the class a.” For example： “Diana is a talkative dinner guest.” In this sentence， “Diana” is the Carrier， and the Attribute is “a talkative dinner guest”.

In the descriptive kind of Attributive relational process， the Attribute is a quality or epithet ascribed to the Carrier， i.e. “X carries the attribute a” . In this kind of relational process， the Attribute is typically an adjective， like the clause “You are very skinny”， here “skinny” was used as Attribute， and it is obviously an adjective.

2. Identifying Relational Processes

An identifying clause is not about ascribing or classifying， but defining. The meaning of an Identifying relational process is that “X serves to define the identity of y.”

e.g. You are the skinniest one here.

In this clause， “you” are identified as the “holder” or “occupant” of the identity or label of “skinniest one”.

You （Identified/Token） are the skinniest one （Identifier/Value） here.

The skinniest one（Identified/Value） here is you（Identifier/Token）.

Sometimes the Identifying and Attributive relational processes seemed similar to each other， but we still have to distinguish them. Take the sentence “the best known of Hangzhou is the west lake.” for example， it looks at first glance much the same as attributive process clauses. However， one significant difference between these clauses and attributive clauses is that they can be reversed， like “the west lake is the best known of Hangzhou”. Attributive clauses cannot normally be reversed in this way. The reason for this is that clause like “the west lake is the best known of Hangzhou” do not simply assign an Attribute to a Carrier. They identify one participant by equating it with another participant. This is why they are called identifying process clauses. And if participant A equals participant B， then of course participant B also equals participant A. In other words， the clauses are reversible. Besides， unlike Attributes， Identifiers cannot be realized by adjective groups. They must be noun groups or nominal clauses. If the second participant is a nominal group， we can decide according to definiteness：  an Attribute is typically indefinite（a/an）， whereas an Identifier is typically definite（the）. The only apparent exception to this is when the Identifier is realized by a superlative adjective. What is more， They can be probed by asking different questions like： “What is X （the Carrier） like？” or “What/Which is X （the Identified）？” While the first one probes attributive processes， the second one probes identifying processes.

There is a strong tendency for the Identifier to be the second participant. This is because the Identifier contains the new or important information in the clause and such information normally comes at the end of the clause. However， sometimes the Identifier can be the first participant. When the Identifier does come first， it is usually marked， in speech at least， by carrying a major pitch movement.

【作者簡介】谭炯，杭州万向职业技术学院。