The following post is by Dr. Kevin Grasso, a linguist and biblical scholar. Dr. Grasso have a BA in Communications from UGA, an MA in linguistics from Dallas International University, a PhD in Hebrew from Hebrew University, and an MA in comparative religion from Hebrew University. He also created a software program called Biblingo, a language learning app to help anyone and everyone learn the biblical languages using modern second language acquisition methods and technology (think Duolingo or Rosetta Stone for the biblical languages). Check out Biblingo’s YouTube channel and The Biblical Languages Podcast for more of Dr. Grasso’s work. -P.S
One of the first reasons I got into linguistics is because I recognized that many of the theological arguments boiled down to the meaning of certain Greek and Hebrew words. I reasoned that if I wanted to contribute to those discussions in the most meaningful way possible, I needed to know Greek and Hebrew. I then reasoned that if I wanted to use Greek and Hebrew responsibly, I needed to have a theory of meaning. Ultimately, I was led to the world of semantics where scholars who don’t care about the theological arguments simply try to figure out how meaning is constructed in language.
Let me be upfront about my theological stance on the egalitarian/complementarian issue. Over my own journey as a follower of Jesus, I have had to revise my thinking multiple times after realizing that I believed something because of a presupposition I had and not my own research. Sometimes the research has confirmed my presupposition and at other times it has made me revise my thinking. I will say upfront that I have not done my homework on all the complexities involved in this issue to have a definitive stance, so when people ask me which camp I fall into, I generally try to avoid being put into either box.
Like many other debates, part of this theological issue (and it is only part) depends on what the word κεφαλή means and how it is used in certain passages in Paul. Preston has already laid out all the data in a very helpful way, so I will not rehash that here. Rather, I want to offer some thoughts on how linguists would approach this problem. In other words, I want to present a theory on the meaning of words that can help us think through the issues involved. To make the discussion more concrete, I will conclude with an exegesis of Ephesians 5:23 using these principles.
My refrigerator is running (italics are used in linguistics for words or sentences that are the object of analysis). Native English speakers do not struggle to understand a sentence like this, even though run is used metaphorically. Despite the jokes this sentence has led to, run here does not refer to the physical action of running, but to a machine that is powered on and functioning. Language is economical in that we regularly reuse words by expanding the meaning to refer to multiple concepts that are distinct but related. Words are often used in a non-literal sense (that is, as metaphors) to refer to distinct things or events. A word like run is notorious for its many, many metaphorical meanings: my refrigerator/nose/sister is running all mean different things. We will come back to this, but when we say my refrigerator is running, we do not ordinarily think of the below image.1Taken from the following site We only think of the sense of run that has to do with a machine being on and functioning properly.
The interpretations of run with refrigerator and nose can be called conventionalized metaphors.2The distinction between conventionalized and novel metaphors discussed here is essentially the same as dead (conventionalized) and live (novel) metaphors. These terms are more often found in the literary literature. They are embedded within the English language, and everyone recognizes that run can have these different meanings. We also use words metaphorically in a one-off instance in what are called novel metaphors.3For discussion, see Kroeger (2018:97-99). I refer to Kroeger’s work throughout because he is a well-respected linguist (PhD from Stanford) who has written an accessible introduction to semantics that is open access, called Analyzing Meaning. He also happens to be the first semantics teacher I had. These metaphors do not have conventionalized meanings. They have meanings that are dependent upon a specific context and make use of the fact that we can use just about any word metaphorically by constructing a context where we embed a word with a particular meaning. For example, Lai et. al. (2009:147) compare all kinds of interesting things that happen in our brains when we encounter these two types of metaphors in the sentences Her life has a new direction and Their style has a new direction. The former is a conventionalized metaphor because life has a sense that is conventionally related to a path of some kind, while the latter is a novel metaphor because style does not have a sense that is related to a path. Given the right context, we can invest style with a path-like meaning, but this is not conventionally associated with the word.
Both of these kinds of metaphors will be important for us as we think about the meaning of κεφαλή ‘head’ (I’ll use single quotes for a translation from Greek into English).
All of this has been an introduction to how we should think about the different kinds of interpretations that a word can encode (or in other words, the kinds of interpretations that a word may contribute to the context). When it comes to analyzing a word like κεφαλή, we have two basic questions: 1) What are the possible interpretations of the word? and 2) What are the features in the surrounding words that would lead us to interpret κεφαλή in a particular way in a given context? To put it differently, (1) what are the options for what κεφαλή can mean, and (2) how do we determine which meaning is correct?
A polysemous word is a single word with multiple related senses. A sense is a particular meaning of a word that has been conventionalized. Going back to the word run, we can see that it can be interpreted differently when we change the thing doing the running. We can label the three senses of run as follows (with definitions only based on our examples):
Sense 1: (of sisters) to move your legs at a fast pace where both legs are never on the ground at the same time
Sense 2: (of refrigerators) to be on and functioning properly
Sense 3: (of noses) to have liquid dripping out of something
About 80% of words are polysemous, so this is a very common language phenomenon (Rodd 2020:411). Sense 2 and 3 are both conventionalized metaphors whereas sense 1 might be called the literal meaning of run. It is not the case that senses are only either “literal” or “conventionalized metaphors,” but those are the two categories of senses that are important for our purposes.4Some words have senses that refer to two different logical types of things. For example, lunch can refer to a physical object (something eaten) or an event (something protracted in time). Neither are metaphorical, but both are legitimate senses (and it is difficult to tell if one is derived from the other). For an important, though highly technical, discussion on this phenomenon, see Asher (2011:ch.2).
At this point, it may be helpful to clarify some of our terms. We have defined sense as a particular meaning of a word. It is a conventionalized way of interpreting a word which the speech community would accept as a legitimate meaning of that word. Whenever I use the word “interpretation,” I am either referring to a sense of a word or a way of understanding a word in a particular context. The word “meaning” is quite flexible, and it is commonly used in one of three ways. First, it may refer to the understanding of a word in a particular context, very similar to the second use of “interpretation.” Second, it may be used as essentially equivalent to “sense.” We can thus speak of a word having multiple meanings, or senses, that are possible contributions that the word makes to the context. Third and finally, “meaning” in linguistics is sometimes used to refer to the contribution of a word regardless of the context. We might also call this the “invariable meaning,” and I do not intend this sense of meaning unless I use “invariable” with it. Although determining the “invariable meaning” is an important part of doing semantics, I do not delve into this issue in any detail in this post.5Some are uncomfortable with the idea of an invariable meaning. There are a number of philosophical reasons why something like this must be the case, and we certainly cannot get into all of the details here. Angelika Kratzer, one of the foremost semanticists of our time, describes discovering the invariable meaning as integral to doing semantics: “It is the task of semantics to describe all those features of the meaning of utterances of linguistic expressions which stay invariable in whatever context these expressions may be used. This invariable element we may call the meaning proper of a linguistic expression” (1977:337). For those who find diagrams helpful, here’s one illustrating the relationship between how I am using “sense,” “interpretation,” and “meaning” and the relationship between the three.
For a word like κεφαλή, we must establish the possible senses, or meanings, of the word. As Preston has mentioned, scholars often suggest four different senses for κεφαλή:
Sense 1: ‘head’ (physical)
Sense 2: ‘leader’
Sense 3: ‘source’
Sense 4: ‘prominence’
We will take a closer look at the examples for sense 3. Before diving into the data, however, we must note a complicating factor in our analysis. Each of the English words that we have used to approximate the different senses of κεφαλή are themselves polysemous. For example, leader is given the following 4 definitions (among others) by Google:
Sense 1: person in charge of a person or group of people
Sense 2: organization/company that is most advanced in an area
Sense 3: principal player in a music group
Sense 4: shoot of plant at apex of stem
When we say that κεφαλή in sense 2 can mean leader, we do not mean that it can mean all of the senses of leader. Rather, we normally mean that κεφαλή maps onto a particular sense of leader, namely leader (sense 1). Just because κεφαλή might be best approximated by leader in contexts that have to do with a person in charge (sense 1) does not mean that κεφαλή would also be used to refer to a “principal player in a music group” (sense 3) or a “shoot of plant at apex of stem” (sense 4). Thus, each sense of κεφαλή must be associated with a particular sense of the English glosses. We can map this out with the diagram below. Throughout our discussion, I will refer to the content of this diagram in order to link senses of English words with senses of Greek words. For example, when I refer to source (sense 1), I am referring to the English word source that is defined as ‘place, person, or thing from which something comes.’ When I say source (sense 2), I am referring to English source as ‘spring from which a river issues,’ and so forth.
Linguists have noted that it can be highly problematic to use English as a metalanguage because it is difficult not to let the meaning of English words creep into our analysis (see Kroeger 2018:7-8 for some discussion of the issues). The trouble is that examples like the second sense of κεφαλή, ‘leader,’ might be best approximated by the first sense of English leader, but that does not make κεφαλή and leader absolutely equivalent even in those contexts. In order to reduce confusion and increase precision, linguists will often use small caps to refer to certain senses, which can then be defined however we want. For example, we might define the second sense of κεφαλή as LEADER “being a social authority of some kind.” We could then ascribe a feature (which is just a component of a word’s sense/meaning) to LEADER, such as [+SOCIAL AUTHORITY]. What we have just done is divorced sense 2 of κεφαλή from English leader and instead made it equivalent to LEADER meaning anyone who is a social authority (and whatever other features we might deem necessary). We will touch on these things in more detail below, but I introduce these concepts here to help get us away from letting English words dictate how we understand the Greek text. Accordingly, I will now use small caps when referring to possible senses of κεφαλή (I still use single quotes for translations into English and italics when discussing the meaning of English words).
In order to think through the meaning of κεφαλή, let’s look at the purported examples of the SOURCE sense to determine which sense of English source maps onto κεφαλή. It could be the case that κεφαλή maps onto more than one sense of source, but just because one sense of source fits one sense of κεφαλή does not mean that κεφαλή can map onto any sense of source. Our basic question is whether κεφαλή maps onto English source (sense 2) or source (sense 1). Here are the examples for κεφαλή with this meaning (from Preston’s third post):
Example 1: From the sources (kephalai) of the river Tearus flows the best and fairest of all river waters. (Herodotus, Histories 4.91)
Example 2: No river that comes from a single spring is smaller at its head (ten kephalēn) than it is thereafter (Galen, On the Doctrines of Hippocrates and Plato 184.108.40.206)
Example 3: For covetousness is the head (kephalē) of every kind of sin (Life of Adam and Eve 19:3).
Example 4: …of all the members of the clan here described Esau is the progenitor, the head (kephalē) as it were of the whole creature. (Philo, Prelim. Studies 61)
In the first two examples, the source of a river is clearly being referred to. This would be sense 2 of source. Of course, this sense is entirely irrelevant for the discussion because no one argues that Christ being the head of the church means that Christ is the beginning of a river. With just these two examples, we could define SOURCE (the sense of κεφαλή) as having three features, such as [+LOCATION], [+START], and [+WATER]. Of course, we would ideally want to be more precise about what these features are, but this suffices for our purposes. Every instance of κεφαλή with the meaning SOURCE has these features. Our question now would be do we have other examples (such as 3 and 4) that would fit this meaning or would force us to change the features of this meaning? Can we say, for example, that SOURCE can actually refer to anything that is [+LOCATION] and [+START] but need not have the water component? Can we say something like Go to the κεφαλή of the race meaning the place where the race starts? We would then revise our features of SOURCE to fit such examples, or we would create an entirely new sense if we could not account for examples 1-4 with the same set of features.6Interestingly, we actually have the same metaphor in English head that Greek uses with κεφαλή. With just the examples from 1 and 2, we could replace the English gloss with headwater. This would restrict the sense of κεφαλή to only water contexts (as our features require for our current meaning of source). It would also reflect a similar metaphor we see in English: head is used in headwater to refer to the beginning of a water location just as κεφαλή is used either to refer either to a literal head of a body or to the beginning of a water location. At the very least, this suggests that a better translation of κεφαλή in these examples would be ‘headwater’ rather than ‘source,’ since it allows us to translate the conventionalized metaphor we find in Greek with a very similar metaphor we see in English.
We are left, then, with examples 3 and 4 to determine whether κεφαλή can mean something close to English source (sense 1). Let’s look at these more closely, starting with example 4. In Greek, the sentence starts with κεφαλή in the phrase κεφαλὴ δὲ ὡς ζῴου which is translated ‘the head as it were of the whole creature.’ Because the word κεφαλή is followed by ζῴου in the genitive, this example is, in my opinion, best understood as referring to the literal sense of ‘head.’ A head of a creature is being referred to. In this context, of course, the word ‘creature’ is being used in an extended metaphor to talk about Esau’s family, but within the phrase itself, κεφαλή simply refers to the literal head. This is invested with new meaning from the context, but the metaphor here is a novel metaphor rather than a conventional metaphor. The novel metaphor is that the creature (ζῴου) refers to Esau’s descendants, and the head (κεφαλή) refers to the thing from which Esau’s descendants come. The word κεφαλή could be translated as ‘source’ here because it refers to the literal top of the creature, namely the head, which is said to be the source of the body in the novel metaphor. However, κεφαλή itself does not seem to be contributing the meaning of source (sense 1) as a conventionalized sense. It just refers to the literal head of a creature, though this is given new significance in the context of the metaphor. At the very least, we would want more evidence to establish this as a legitimate sense of κεφαλή.
Finally, we come to example 3. In this example, desire is said to be the κεφαλή of all sin. As Preston discusses in a footnote, this example is debated with ‘source’ (which would mean something like cause?) and ‘beginning’ as the main translation options. LSJ lists other examples where ‘beginning’ seems to be the meaning (where we have things like κεφαλὴ χρόνου ‘head/beginning of time’), so this is an attested meaning outside of this passage. The difficult thing about this example is that several potential glosses could fit the context. Desire could be the cause of all sin, the beginning of all sin, or even the chief/greatest of all sin—all of which would make sense. How do we decide? Unfortunately, this is one of those cases where it is very difficult. There aren’t enough triggers in the context to help us definitively determine the meaning (more on this below), though my analysis would be that the context lends itself best to a temporal interpretation. All sin starts at desire seems to be what is being communicated. At the very least, this is not enough evidence for me to conclude that κεφαλή is contributing the meaning SOURCE (as cause), particularly considering the fact that we have clear examples where beginning must be the meaning, and that sense is at least as good as source (as cause) here. We could then have a new sense of κεφαλή that we could call BEGINNING with the features [+TIME] and [+START].
This whole discussion has simply been to establish what our options are for κεφαλή. Based on my reading of the data, I don’t think κεφαλή ever means source (sense 1). I would argue for source (sense 2) and then an additional temporal sense meaning beginning (of an event). We now turn to how we can determine which sense is the best interpretation in a given context.
How polysemy is resolved
Let us return to run. We noted three senses of the word run, and we gave different contexts where run is interpreted differently. These examples give us several clues as to how to resolve ambiguity. First, we should remember that we interpret run to mean only one sense of run in any given occurrence of the word (think again of the “running” refrigerator—it is not that the refrigerator is functioning and moving quickly on legs at the same time). This is generally how polysemy works (see Kroeger 2018:94-96 for discussion).7There is a phenomenon called zeugma where two senses of a polysemous word are activated simultaneously. For example, we can say something like My sister and my refrigerator are running. In this case, running has to be interpreted with two different senses because of the different arguments it takes, but cases like these are very rare and are really not relevant to the discussion. Second, we noted in our definition that the senses were usually talking about different things. For a sister to run is different than for a refrigerator to run, and both differ from noses. In other words, each sense of run is triggered by certain words in the context. Given our definitions, we can predict the sense of run when sister, refrigerator, and nose are the subjects. With more data, we could also begin to make more predictions. For example, the words brother, computer, and faucet can be used with the three senses of run we have described. For the meaning of run as moving one’s legs, we could say that some feature like [+SIBLING] activates sense 1 of run if we saw that both sister and brother triggered that interpretation. Of course, we know this would not capture all the data—the feature [+SIBLING] is also too narrow of a category. With more and more data, we could refine what it is that triggers sense 1 of run, and we would probably come up with something like [+ANIMATE]. Animate things, such as animals or people, engage in running according to sense 1. If this is correct, it allows us to predict when we have sense 1 of run activated: when the thing doing the running is animate (i.e. has the feature [+ANIMATE]), we have sense 1.
Crucially, the above discussion is not a vague appeal to context in order to determine the correct interpretation of a word. Polysemous words must be disambiguated, which means that we must ascribe a particular sense to each use of the word. Whichever sense is given the correct interpretation in the context can be said to be “activated.” In the sentence Your refrigerator is running, the sense activated is the ‘functioning’ sense, and your refrigerator is the primary disambiguating cue, particularly the fact that it is a machine. The reason why the joke Then you’d better go catch it is indeed a joke is because the sense of run as moving one’s legs is not activated—we know that to be an interpretation not intended by the speaker. Although sense disambiguation is a complex phenomenon, the most important disambiguating cue is the immediate sentence context (Rodd 2020:412), and even the most immediate thing that the word combines with. The word run, for example, combines directly with my sister/refrigerator/nose, and it is the meaning of these words that forces us to interpret run in the way we do in any given context.8I would refer those with a high degree of interest (and theoretical background) to the work of Asher (2011) in this respect, whose work on how predication and lexical meaning interact is foundational.
I hope the utility of this discussion on run is obvious, but let’s spell it out. In order to determine what sense of κεφαλή is activated in a given context, we have to figure out what features are associated with each of its senses, and then we have to look carefully at the context to determine if any of those features are activated. To illustrate, we might say that the literal sense of κεφαλή is activated when we have the feature [+BODY] in a word that combines directly with κεφαλή. We saw this in κεφαλὴ…ζῴου ‘head…of the creature.’ Because creatures carry the meaning component of [+BODY], we interpret κεφαλή “literally.” For a sense like LEADER, we might have the feature [+SOCIAL HIERARCHY]. This would mean that whenever we have κεφαλή combining with a word that has this feature activated in the context, we would expect this meaning. If SOURCE as “cause” were a true sense, we would expect to find this sense where we have a feature of something like [+EFFECT] in the thing that κεφαλή is combining with. The word κεφαλή would then refer to the cause. As we said, though, we have clear examples where κεφαλή combines with something that is [+TIME] to mean BEGINNING, but other senses of English source (outside of the headwater of river sense) are dubious at best.
Let us return to our example in the Life of Adam and Eve before turning finally to Ephesians. What makes the example difficult is that we are not sure if ‘desire’ is being conceptualized as the beginning of sin or the cause of sin. In theory, the word ‘desire’ by itself could be either of these. The greater context would have to help us disambiguate how ‘desire’ is functioning, either as an indicator of time or cause, but it is hard to tell how it is being conceptualized.9One could argue that either feature is present in the context. Part of the trouble is that causes always precede their effects, so it is sometimes hard to tease apart the difference between one element that is causally related to another and one that is temporally related. If κεφαλή did not have the meaning of SOURCE as a cause, then Greek speakers would have intuitively thought of time (since we know that that was a legitimate sense). Perhaps the sense of SOURCE (as cause) was also an option, in which case this kind of example might have been genuinely ambiguous. Unfortunately, we don’t have the data to support this in the absence of clearer examples of the SOURCE (as cause) meaning.
We now have the basic tools we need to analyze Ephesians 5:23 and to try to disambiguate the text. Because the verse starts with a causal conjunction, we would want to go back to at least 5:21-22 of which 5:23 provides the reason. In 5:21-22, we read ‘submitting to one another in the fear of Christ, wives to their own husbands as to the lord…’ The basic question we would want answered about the participants here is how they are being conceptualized. Are sources/causes and products/effects being talked about, or are social hierarchies being discussed? If it is the former, the word ἀνήρ ‘man’ in 5:23 could be construed as a ‘head’ meaning SOURCE (if that were a legitimate sense of κεφαλή). If it is the latter, ‘man’ would be construed as being part of a social hierarchy, of which he would have a certain relation, namely as the κεφαλή meaning LEADER. Given this way of framing the question, the key to our analysis is actually found in the verb ὑποτάσσω. What component of meaning is this verb activating that would lead us to interpret ἀνήρ and Χριστός as either people that are sources or people that are social authorities? When used of people, as it is in this context, this verb really only means being subjected to someone as in submitting to them (e.g. Luke 2:51; Tit 2:9; 1 Pet 2:18; etc.). Given this, I would say that the words γυναικός and ἐκκλησίας (the words that κεφαλή immediately combines with) inherit something like the feature [+SOCIAL HIERARCHY] from the verb ὑποτάσσω, and this disambiguates the sense of κεφαλή to mean LEADER (i.e. someone with social authority).
A related text to Ephesians 5:23 is Colossians 1:18. The contrast between these two is instructive. In Colossians 1:18, we have καὶ αὐτός ἐστιν ἡ κεφαλὴ τοῦ σώματος τῆς ἐκκλησίας ‘And he is the head of the body, the church.’ In this case, the word κεφαλή combines immediately with σῶματος ‘body.’ The word σῶματος is itself being used metaphorically to refer to a people, but the word κεφαλή refers to a physical head of the metaphorical body. Bodies have heads, so that is what is being referred to. In the context, the words κεφαλή and σῶματος are invested with different meanings as part of an extended metaphor about Christ and the church, but these kinds of meanings would fall under the category of novel metaphor (as described above, this is a one-off metaphor with a meaning dependent upon a specific context). They are not meanings that κεφαλή is itself contributing. It just means literal head within the metaphor. This is very similar to the example we saw in Philo above.
In contrast to this, Ephesians 5:23 has κεφαλή combining directly with γυναικός ‘woman’ and ἐκκλησίας ‘church/gathering.’ While a ‘woman’ has a physical head (and so passes the [+BODY] feature test), it is impossible to interpret ἀνήρ ‘man’ as the literal head of that body. Moreover, ‘woman’ in this context is also part of a list of words that all have social relations, particularly as contributed by the verb ὑποτάσσω. This suggests that it is this feature [+SOCIAL HIERARCHY] that is most salient in the context. When we move to ‘church/gathering,’ this becomes unambiguous. Churches/gatherings don’t have physical heads or bodies. They are a social group with a social function that is being commanded in this context to have a particular social orientation, namely to ‘submit’ (ὑποτάσσω) to one another (5:21). This makes the meaning of κεφαλή necessarily LEADER [+SOCIAL HIERARCHY], since the thing that it combines with only has this feature and not others.
Much more could be said, but I hope that this discussion provides a different way of thinking about the problem. Again, I do not have a dog in this fight , and my main concern is not proving one side or the other but to provide tools, so that we can all read the Bible responsibly and submit to it. As much as is possible, we must read the text with hearts and minds open and then build our theology off the text, not the other way around.
Based on my reading of the data, I doubt that the SOURCE sense even exists in the way that scholars describe it. It is clear that κεφαλή can mean ‘headwater’, but it is not clear that it can mean anything broader than this that would map onto English source. I do think BEGINNING would also be a clear sense of κεφαλή. However, this just gives us what options are out there for how κεφαλή is used. It does not tell us how it is used in a particular context. In Ephesians 5:21-24, the verb ὑποτάσσω is the disambiguating cue that leads me to interpret κεφαλή as referring to someone who is part of a social hierarchy, as the English word leader suggests, though other glosses may also be possible. The most obvious one, and what I would opt for, is ‘head,’ which can also be used to refer to a social authority and which is a pretty neutral term in and of itself (the “head of an organization” is neither bad nor good—just the person with the most authority in that organization). On this note, we must beware of thinking that κεφαλή = leader here or even κεφαλή = head. It does not. The word κεφαλή = κεφαλή. The trouble is that we have all kinds of baggage with the words leader and head in English (particularly in this debate), and we carry this into our interpretation when we read that word. It might not be the case that κεφαλή has the same baggage. At the very least, this analysis does not answer what kind of κεφαλή the man is to be to his wife or the Messiah is to be to His church. For that, one must continue to read Ephesians 5:25 and following. In my reading, such a leader is tasked with a very high calling to love and submit oneself to the good of the other.
- 1Taken from the following site
- 2The distinction between conventionalized and novel metaphors discussed here is essentially the same as dead (conventionalized) and live (novel) metaphors. These terms are more often found in the literary literature.
- 3For discussion, see Kroeger (2018:97-99). I refer to Kroeger’s work throughout because he is a well-respected linguist (PhD from Stanford) who has written an accessible introduction to semantics that is open access, called Analyzing Meaning. He also happens to be the first semantics teacher I had.
- 4Some words have senses that refer to two different logical types of things. For example, lunch can refer to a physical object (something eaten) or an event (something protracted in time). Neither are metaphorical, but both are legitimate senses (and it is difficult to tell if one is derived from the other). For an important, though highly technical, discussion on this phenomenon, see Asher (2011:ch.2).
- 5Some are uncomfortable with the idea of an invariable meaning. There are a number of philosophical reasons why something like this must be the case, and we certainly cannot get into all of the details here. Angelika Kratzer, one of the foremost semanticists of our time, describes discovering the invariable meaning as integral to doing semantics: “It is the task of semantics to describe all those features of the meaning of utterances of linguistic expressions which stay invariable in whatever context these expressions may be used. This invariable element we may call the meaning proper of a linguistic expression” (1977:337).
- 6Interestingly, we actually have the same metaphor in English head that Greek uses with κεφαλή. With just the examples from 1 and 2, we could replace the English gloss with headwater. This would restrict the sense of κεφαλή to only water contexts (as our features require for our current meaning of source). It would also reflect a similar metaphor we see in English: head is used in headwater to refer to the beginning of a water location just as κεφαλή is used either to refer either to a literal head of a body or to the beginning of a water location. At the very least, this suggests that a better translation of κεφαλή in these examples would be ‘headwater’ rather than ‘source,’ since it allows us to translate the conventionalized metaphor we find in Greek with a very similar metaphor we see in English.
- 7There is a phenomenon called zeugma where two senses of a polysemous word are activated simultaneously. For example, we can say something like My sister and my refrigerator are running. In this case, running has to be interpreted with two different senses because of the different arguments it takes, but cases like these are very rare and are really not relevant to the discussion.
- 8I would refer those with a high degree of interest (and theoretical background) to the work of Asher (2011) in this respect, whose work on how predication and lexical meaning interact is foundational.
- 9One could argue that either feature is present in the context. Part of the trouble is that causes always precede their effects, so it is sometimes hard to tease apart the difference between one element that is causally related to another and one that is temporally related.
Asher, Nicholas. Lexical meaning in context: A web of words. Cambridge University Press, 2011.
Kratzer, Angelika. “What ‘must’and ‘can’must and can mean.” Linguistics and philosophy 1.3 (1977): 337-355.
Kroeger, Paul R. Analyzing meaning: An introduction to semantics and pragmatics. Language Science Press, 2018.
Lai, Vicky Tzuyin, Tim Curran, and Lise Menn. “Comprehending conventional and novel metaphors: An ERP study.” Brain research 1284 (2009): 145-155.
Rodd, Jennifer M. “Settling into semantic space: An ambiguity-focused account of word-meaning access.” Perspectives on Psychological Science 15.2 (2020): 411-427