Analyzing the lexical meaning of τελειο- in 1 John

The problem

The following three popular English translations represent different (contradictory) ways of understanding John’s message in 1 John 4:18:

Such love has no fear, because perfect (τελεία) love expels all fear. If we are afraid, it is for fear of punishment, and this shows that we have not fully experienced (τετελείωται) his perfect love. (NLT)

A real (τελεία) love for others will chase those worries away. The thought of being punished is what makes us afraid. It shows we have not really learned (τετελείωται) to love. (CEV) There is no fear in love, but perfect (τελεία) love casts out fear. For fear has to do with punishment, and whoever fears has not been perfected (τετελείωται) in love. (ESV)

These three translations do not merely represent three equally valid ways of rendering the same meaning into English, but they reflect different understandings of the underlying Greek text. In the NLT, the ‘perfect love’ that expels fear is explicitly said to be ‘his’ (God’s) perfect love (the pronoun ‘his’ is lacking in the Greek). According to this understanding, if we have ‘fully experienced’ his perfect love, we would have no reason to fear being punished, so it is God’s love that causes us not to fear. The CEV translation, however, reads the Greek differently – it is a ‘real love for others’ that causes us not to fear, and what was translated as ‘fully experience[ing] his perfect love’ in the NLT is rendered ‘we have … really learned to love,’ making the love ours and not God’s. John is either saying that God’s love makes someone not fear punishment (so the NLT) or the believer’s love makes him not fear (so the CEV), but he is not saying both. In contrast to both, the ESV is a more “literal” translation that leaves the text more opaque (as do many other English translations, such as the NIV, NASB, KJV, etc.). The question is what it means to have been perfected in love. If we reason that humans cannot be ‘perfect in love’ in the sense that they love perfectly, then we might think that the ‘perfect love’ being referred to is God’s love (following the NLT). The trouble is that if the perfect love that casts out fear must be God’s (since he only has perfect love), then what does it mean for a person to be perfected in love at the end of the verse? This leads to two logical options for the ESV: either everyone fears punishment (since no one has perfect love) or some people really do have perfect love (and, therefore, have no need to fear punishment). Because both of these options are unlikely, we might be tempted to think that somehow the perfect love is God’s love, leading us back to the NLT understanding. As we have said, the best description of the ESV is opaque.

These translation issues reflect different exegetical stances taken by the translators. The main issue is the meaning of the two Greek words that come from the stem τελειο-. In the NLT and ESV, these are translated ‘perfect.’ From this, the NLT translators seemed to reason that the ‘perfect love’ being spoken of must be God’s love, since only God’s love can be perfect. The ESV translators, on the other hand, agree with the translation of ‘perfect’ for these words, but the one who has been made perfect in love seems to be the one who does not fear judgment, i.e. a person, resulting in the logical problem described above. The CEV takes a different route. The translators’ use of the word ‘real/really’ rather than ‘perfect’ reflects a different understanding of our two key Greek words, which allows it to readily apply to people. In this exegetical section, we will focus on the verb τελειόω to illustrate how we can use tools in lexical semantics to help us solve this exegetical problem.

Before we get into the data, it is worth mentioning at the outset that this problem concerns the meaning of English words just as much as Greek words. When we gloss a Greek word, we are saying that such a word is best approximated with a certain English word and that the two words share semantic features. For example, the word perfect is very commonly found in a phrase like Nobody’s perfect, and the first definition on Merriam-Webster’s dictionary online is “being entirely without fault or defect.”1 This does not necessarily tell us the meaning of the word; a definition is only as good as it accurately reflects a word’s distribution. However, it may, in fact, reflect the distribution of perfect, and even if it did not, it tells us something about how people commonly think about the word – it refers to an ideal to which humans cannot normally attain. This meaning of ‘without flaw’ is precisely the issue we discussed above as causing the NLT translators

(possibly) to assume that the perfect love must be God’s love. Our question is, even if the Greek words with the τελειο- stem best approximate English words with the perfect stem, does this meaning component carry over? And if it does, does it always? If not, in what contexts do we find it? These are the main problems we will address.

The data

The following chart represents my own study of the Greek verb τελειόω. As can be seen by the glosses, I take the stance that there are no unambiguous examples of the verb τελειόω meaning ‘to make perfect,’ as in to make morally flawless. While making someone perfect in English normally has this meaning, τελειόω in Greek does not mean this, which is my central claim and what we will work towards demonstrating in this section.

Table 109 (continued)

These data represent the result of having gone through all the examples of τελειόω and having determined the sense of the word in each passage. In the rest of this section, we will walk through how I made the table and why I glossed the senses in the way I have. In the steps presented in Semantics A, we said that the first thing to do when collecting the data is to organize it with relevant information. The columns in the above table show the most relevant properties for analyzing the meaning of verbs. In our discussion on the lexical semantics of verbs, we said that the two main properties are temporality and argument structure. We will focus on the latter. This involves not only how arguments are structured in the syntax, but also what kinds of arguments the verb may select for. Hence, I have tagged the entries in the subject, complement, and adjunct columns with thematic roles in addition to giving the Greek word that occurs there. A thematic role refers to the way in which an argument participates in the eventuality. It is not a description of the properties of the participants, but it describes what kinds of requirements the verb places upon its participants. For example, the two arguments in Selah bonked Laura have the same basic properties of being volitional, sentient humans, but they do not have the same roles in the event of hitting. We would label the subject Selah an agent who is intentionally engaging in the action described by the verb, but Laura would be a theme who is the entity that is being acted upon by the agent. Because the two arguments have the same inherent properties, it is the verb bonk that specifies the presence of certain meaningful components in its subject and object. These meaning components are what we are trying to capture when we group the arguments into thematic roles. There have been many of these kinds of lists proposed about how entities may participate in events, and there are theoretical problems with trying to do this for every verb. Nevertheless, it is helpful to group uses into general buckets.

The most important principle when engaging in this analysis is to choose the most appropriate grain-size. I say “appropriate” and not “correct” because there is not necessarily a correct answer here. The grain-size will determine, however, the explanatory power of the analysis. The goal should always be kept in mind: when we label an argument as having a certain semantic role, we are trying to determine meaningful differences between arguments to see how those differences affect the interpretation of the verb. This takes careful thought, and this is one reason why referring to a list of roles can be helpful. An example of being too fine-grained from the world of Hebrew lexicography is the Dictionary of Classical Hebrew (DCH). This dictionary gives every single argument found for every verb. For example, every name found that occurs as the subject of a verb is listed. This would be like listing all possible names of who might be able to participate as the subject in the event described by bonk – in other words, we would list every possible name! This is far too fine-grained. If Selah can serve as the subject of bonk, so can Laura, since they can both be agents. In such cases, it is far more helpful to show that an agent may serve as the subject of the verb. On the other hand, BDAG is not fine-grained enough. The only distinction made with any consistency is between objects that are animate (marked with τινα) as opposed to objects that are inanimate (marked with τι; see 21.2.4.1 for this paradigm and grammar topic).5 An entry like ἐσθίω illustrates how thematic roles can explain different senses. The second sense is “to do away with completely … consume, devour ” (2000:396). This differs from the first sense which is “to take something in through the mouth.” If you look at the subjects of the examples in the second sense, they all involve inanimate entities, whereas the first sense has animate entities like a person. Because only animate entities can be agents, those subjects in the second sense are better described as causer s. The distinction between an agent and a causer is a well-researched phenomenon in lexical semantics, and it explains why we have the two senses of ἐσθίω as well as when to interpret one or the other.

Returning to the data, we start with cases where the sense is obvious, such as John 4:34:

• (106) τελειώσω αὐτοῦ τὸ ἔργον

I will accomplish his work

This example demonstrates that ‘to perfect’ is certainly not a viable understanding in every context. When an agent is the subject and an event is the complement, the verb means to accomplish/finish/execute that event. This holds in every occurrence with these kinds of arguments.

Having established that τελειόω does not always mean ‘to perfect,’ the next question to address is whether it ever means this, preferably outside of the contested passages. Examples where the event described by the verb involved the moral perfection of a person would be ideal. A possible candidate for this is Heb 9:9:

• (107) δῶρά τε καὶ θυσίαι προσφέρονται μὴ δυνάμεναι κατὰ συνείδησιν τελειῶσαι τὸν λατρεύοντα
• Gifts and also sacrifices being brought are not able to satisfy the worshiper with respect to conscience

There are three elements in the event described by τελειῶσαι in this example. First, there is something that is doing the action, which may be an instrument, agent, or causer. In this case, gifts and sacrifices are best interpreted as instruments. Next, there is someone who is affected by the action, namely the ‘worshiper,’ which we have labelled the theme. Finally, the theme is affected with respect to a certain property. It is not that the worshiper is generally affected, but a specific property in them is, namely their conscience.

Although some English versions translate τελειῶσαι as ‘perfect’ (ESV, NASB, KJV, and many other more “literal” translations), the sense is not that the worshiper becomes morally perfect. Rather, it is the property of the worshiper that is affected by the action of the verb, not the worshiper himself/herself. Because we do not normally speak of a conscience being “perfected,” other translations like the NIV and NLT opt for ‘clear the conscience’ or ‘cleanse the consciences’

respectively. The question under discussion in the passage is about the effectiveness of sacrifices, whether they do what they are supposed to do or not. They are supposed to cause the worshiper’s conscience to be cleared, but according to this verse, they cannot. Despite having a person as the object here, there is no moral transformation, and the only change is specified as being the person’s conscience. The quality that the conscience of the person takes on is that it is cleared. This is implied in the property itself, since the quality that consciences have is either to be clear/innocent or guilty. The verb says that such a quality is manifested.

At other times, the quality itself is named rather than the property:

(108) καὶ ἐκ τῶν ἔργων ἡ πίστις ἐτελειώθη (Jas 2:22)

• And from works faith is manifested

English translations generally either opt for the glosses ‘complete’ or ‘perfect’ in Jas 2:22, but neither of these translations is adequate. Faith is not completed from works, which would suggest that it is an activity that is now over (per the usual sense of ‘complete’), nor is faith perfected from works, which would mean that works caused faith to not have any flaws. Rather, the quality of faith is expressed in the world by works. This is exactly what the context would suggest (and this kind of meaning is consistent with the use of τελειόω to refer to performing a deed as in (106)). James 2:18 explicitly says δείξω ἐκ τῶν ἔργων μου τὴν πίστιν ‘I will show you my faith from my works.’ After giving Abraham as an example of someone who shows faith from works, the more general statement is given in James 2:22 that faith is ἐτελειώθη from works. The best understanding of our word with this argument structure and this kind of context, then, is ‘manifest,’ meaning works manifest someone’s quality of faith. Crucially, this meaning may be expressed by the English glosses ‘manifest’ or ‘express’, but this does not mean that this sense of τελειόω is equivalent to these English words. It just means that these English words best approximate the Greek in most examples with this sense. The goal in a lexical analysis is to come to a better understanding of what the Greek means first and then to ask the related question of how to best represent that meaning in the target language.

Another mismatch between the English gloss ‘to perfect’ and the sense of τελειόω that we have been exploring is the nature of the quality. While the quality manifested is positive in Jas 2:22, it need not be. The Epistle of Barnabas 14:5 demonstrates this clearly:

(109) κἀκεῖνοι τελειωθῶσιν τοῖς ἁμαρτήμασιν

And that those people might be manifested in their sins

In context, this is speaking of the Jews who rejected Jesus as the Messiah. By rejecting him, they are shown to be sinners, and the Christians are shown to be the inheritors of the covenant, according to Barnabas. The English word ‘to perfect’ is impossible here, since ‘perfect’ in English refers to a positive quality.

The solution

Before we come to the verses we began with, consider how the three translations first cited handle 1 John 4:12 where we find the participle of the same verb we have been discussing:8

No one has ever seen God. But if we love each other, God lives in us, and his love is brought to full expression (τετελειωμένη) in us. (NLT) No one has ever seen God. But if we love each other, God lives in us, and his love is truly (τετελειωμένη) in our hearts. (CEV) No one has ever seen God; if we love one another, God abides in us and his love is perfected (τετελειωμένη) in us. (ESV)

Instead of translating our verb as ‘perfect’ as the NLT does in 1 John 4:18, the NLT in this verse says ‘brought to full expression.’ This is the same sense we have been arguing for in other passages and which we demonstrated is found in

James 2:22. We also have the same elements: there is a quality, ‘love,’ that is manifested in a particular people. The ESV again translates this as ‘perfected,’ but we have already said that this does not mean that God’s love is made perfect inside of us – which is what the English would normally mean. Rather, the NLT gets this exactly right, and the CEV, though unnecessarily adding ‘in our hearts,’ gets the general sense correct as well.

When we come back to 1 John 4:18, the NLT suddenly translates τελειόω as ‘experienced perfect love.’ The fundamental issue with this is that the word τελειόω does not specify a certain quality, as we have seen, but the translation suggests that it is God’s “perfect” love that is now experienced. This is probably because the adjective τελεία at the beginning of the verse modifies ‘love’ and has been translated as ‘perfect.’ However, given that the verb and adjective are closely related (the verb is derived from the adjective), it is reasonable to suppose that they refer to the same thing here. We have already said that the verb cannot mean ‘perfect’ and probably means ‘manifested/expressed’ as it does in 1 John 4:12. This is further supported by the fact that the arguments of the verb are very similar in both examples. In both verses, we have a quality expressed as well as a possessor just as we have in (109) where we saw that the quality could be negative. This would lead to the following translation:

(110) Final translation of 1 John 4:18

• φόβος οὐκ ἔστιν ἐν τῇ ἀγάπῃ, ἀλλ’ ἡ τελεία ἀγάπη ἔξω βάλλει τὸν φόβον, ὅτι ὁ φόβος κόλασιν ἔχει, ὁ δὲ φοβούμενος οὐ τετελείωται ἐν τῇ ἀγάπῃ.
• Fear is not in love, but manifested love expels fear, since fear is about punishment, and the one who fears has not been manifesting love

We have gone through the data carefully to demonstrate how to do a word study with modern lexical semantic tools. Let us recap the steps we have taken, so that you can replicate the process with another word. First, we began by asking what the meaning component is in τελειόω that we are questioning. This allowed us to search for this component in the data. Second, we mapped out the data by indicating the semantic role of each argument in all the examples. Third, we connected these similar syntactic patterns with senses of the verb. This allowed us to establish a preliminary set of glosses and to answer our question whether the word τελειόω always means ‘to perfect’ (which we answered negatively). Fourth, we looked at cases where the meaning of ‘to perfect’ is possible and we looked carefully at the role of the arguments described by the verb. We concluded that the word does not actually mean ‘to perfect,’ but rather, it means that a certain quality, whether positive or negative, is ‘manifested’ or ‘expressed.’ Because it says nothing about the nature of the quality (whereas ‘to perfect someone’ typically means that someone has a moral quality that is without flaw), we concluded that the meaning components of ‘to perfect’ and τελειόω do not match in the relevant way. Finally, we looked at other examples where a quality was manifested in a person, and we saw that this most closely matches our verse in question in 1 John 4:18, both syntactically and semantically.

Conclusion

When exegeting a certain passage or doing a word study, there tends to be a heavy bias towards whatever prototypical English gloss is found for a word in the introductory textbooks or lexicons. Hence, BDAG has the sense of ‘make perfect’ with many of the verses we discussed above as examples. There are a number of reasons for this, but it represents a faithfulness to the English tradition of studying Greek and not necessarily faithfulness to the Greek. Word meanings are determined by their distribution, not by glosses within a lexicon. Sometimes lexicons are helpful, and at other times, they can lead us astray. In this case, it is the tradition of the KJV that has led many astray, but the KJV, I would argue, supports our analysis despite the use of the word ‘perfect.’ The KJV says ‘He that feareth is not made perfect in love.’ How is ‘perfect’ used here? The Oxford English Dictionary (OED) lists the first sense of to perfect as “to bring to fulfillment or full development (obsolete).” One of the examples under this sense from 1699 says “To perfyte publickly what he hade formerly essayed privately.” This is very close to what we have suggested for τελειόω in our context. When something is “perfected,” it is manifested or expressed in the world. Even in 1870, we can see other similar examples listed in the OED, such as “And all those images of love and pain, Wrought as the year did wax, perfect, and wane.” For the images of love and pain to “perfect” means that they were expressed or manifested, again exactly the meaning in our context. When the KJV translators used the word perfect, I suggest that they were intending a sense that fit the word τελειόω at the time, but which is now, to the confusion of many, “obsolete” according to the OED. The word τελειόω no longer means ‘to perfect’ in contemporary English.

Suggested exegetical exercise

Analyze one of the following verbs by using a table and filling out the relevant syntactic and semantic information: διώκω, λύω, ἐργάζομαι, ἐνδύω, or βαστάζω. Pick one of the passages where the verb is found that appears to be ambiguous (or is treated differently by different translations). Make an argument for one of the senses of the verb based on your lexical analysis. Suggested length: 3,000 words