Musings on πᾶς
I replaced the picture of my tree diagram for 4.16 HERE. I became a bit frustrated over the structure of the Greek noun phrase with reference to the word πᾶς, so I had to do a bit more studying.
πᾶς does not function like other adjectives in terms of its structure. The typical structure for adjectives is:
The Relation of Adjective to Noun
A. When the Article Is Present (306-309)
1. The Attributive Positions: adjective modifies the noun
a. First Attributive: article-adjective-noun (ὁ ἀγαθὸς βασιλεύς = the good king)
b. Second Attributive: article-noun-article-αδ ̓εχτίε(ὁ βασιλεὺς ὁ ἀγαθός = the good king)
2. The Predicate Positions: adjective makes assertion about the noun
a. First Predicate: adjective-article-noun (ἀγαθὸς ὁ βασιλεύς = the king is good)
b. Second Predicate: article-noun-adjective (ὁ βασιλεὺς ἀγαθός = the king is good)
B. When the Article Is Absent
1. The Anarthrous Adjective-Noun Construction: usually attributive, sometimes predicate
2. The Anarthrous Noun-Adjective Construction: usually attributive, sometimes predicate[1]
Following Palmer’s understanding of the Greek Noun Phrase, the Attributive Position would look like this:The 1st Attributive Position is on the left and the 2nd is on the right. In this case, the adjective and adjective phrase (A & AP respectively) are found within both the NP and the N-Bar. This is roughly equivalent to “the good king” in English.The Predicate Position would be:Again, 1st Position is on the left.Like the name sounds, in the Predicate Position the A & AP are both independent from the NP and function as the predicate of the clause. These types of clauses have no verb. Thus the AP is referred as a “non-verbal” predicate or “the semantic predicate.” The translation equivalent would be, “The king is good.”
If we assume that πᾶς functions from within the Noun Phrase itself, we would have the following structures:
πᾶς. The adjective πᾶς denotes the concept of completeness, whether it specifies a conglomeration of individual parts (each, every) or an undifferentiated whole. When it occurs in predicate structure [i.e. πάντες οἱ ἄνδρες], it specifies the items as considered together, and is often translated ‘all’ (extensive use). When it occurs in attributive structure it specifies completeness and is often translated ‘whole’ or ‘entire’ (intensive use) [i.e. ὁ πᾶς νόμος]. When it occurs in attributive structure without the articular substantive it is often translated ‘each’ or ‘every’ (it may be intensive or extensive). It is often difficult to differentiate this last structure from the two structures above, illustrating that the above schema is not a definitive guide to translation and understanding.[3]
Like the adjective, there are two specific articular structures and also like the adjective, when the noun is anarthrous, it becomes difficult to determine exactly which meaning is intended. Here is the evidence for the structure of πᾶς:
- Semantic/Syntactical Distinction – there are two definite positions that convey different meanings.
- Semantic Ambiguity – when the article is missing from the noun it becomes near impossible to determine which meaning is intended.
- Restricted Distribution – πᾶς cannot simply be placed anywhere in the sentence. It is limited to a narrow area around near the article and noun:

Thus πᾶς would not be found function modifying any of the words below:

More specifically, if πᾶς did occur in such position beside a preposition (which is entirely possible), it would not modify the head noun ἁφῆς or any other word above. Instead it would modify a word from a previous phrase directly adjacent.
On the basis of these facts, I propose the following structure for πᾶς.


These two diagrams do well to represent the data. The attributive position, for the most part, parallels the structure of the adjective and the so-called “predicative position” does well to distinguish the difference in meaning from the attributive. In both cases, the article (D1) in the diagrams helps point out which position πᾶς is functioning in. This explains why the anarthrous noun phrases cause the semantic/structural ambiguity between the two positions.
These structures are extremely consistent throughout the New Testament. For the 1243 occurrences of πᾶς, there are only seven exceptions, which I hope to examine and deal with in the near future. If anyone actually stayed interested in this post to the end, the references are: Mark 3:28; Mark 12:43; Luke 1:65; John 2:15; Acts 26:14; Colossians 4:9; Titus 3:2. I have not taken the time to examine all of these in detail, so it may very well be that they are not actually exceptions at all. But that question will have to wait until next week.
[1]Adapted from Daniel B. Wallace, Greek Grammar Beyond the Basics – Exegetical Syntax of the New Testament (Zondervan Publishing House and Galaxie Software, 1999; 2002), 737.
[2] This is a point that Michael Palmer stated on page 67, but never developed in his book Levels of Constituent Structure in New Testament Greek.
[3]Stanley E. Porter, Idioms of the Greek New Testament (Sheffield: JSOT, 1999), 119.



I am reminded here of the rules for autos, as I remember them from long ago. Effectively this word used as an adjective has two different senses based on its position: when in the normal attributive position, the meaning is “the same”, and when it is in the normal predicative position, the meaning is “self”. It seems that on your analysis pas works in two senses rather similarly, which might tend to confirm your understanding.
Peter Kirk
February 7, 2008 at 5:45 pm
Hmmm, that’s interesting – and makes sense. I might have to look into that.
Have you read Palmer’s book? He works through the structure of οὔτος. He makes a very strong case for the N-Bar on the basis of distribution and semantics. For example, ὀ can substantize prepositional phrases, participles, and just about any other kind of phrase. One of the end results is that ὀ is labeled D1 and ού̓τος is D2.
Mike
February 7, 2008 at 7:55 pm
[...] this week, I presented arguments for a structural representation for the Quantifier πᾶς (HERE). In explaining the function of various words there would be significantly more benefit for [...]
A Plea for Inductive Greek Study « εν εφέσω: Thoughts and Meditations
February 9, 2008 at 9:36 pm
[...] Greek of the New Testament from a modified generative perspective”. This month he discussed the use of pas, predicative and attributive adjectives, and evidence for the Greek noun phrase. And if you are [...]
Blue Cord » Biblical Studies Blog Carnival XXVII
March 3, 2008 at 10:22 am
[...] This is a continuation of a couple things. First of all, this is part of my working through Ephesians. But also this post continues to examine the attributive position in the Greek Noun Phrase, which has already been discussed HERE, HERE, HERE, and HERE. [...]
The Attributive Position in Ephesians 4.18 « εν εφέσω: Thoughts and Meditations
April 12, 2008 at 9:37 pm