God's pronouns

Abstract

The Church of England is currently debating what pronouns to use of God in liturgy. Opinions are strongly established on various sides. This article aims to slow the pace at which strong judgements are arrived at, through four sets of arguments. First, the distinctiveness of English compared with some other European languages and the danger of allowing the contingencies of English pronoun use to dominate the possible meanings of scripture. Second (drawing on the work of Janet Martin Soskice), the complexity of the figure of the fatherhood of God. Third, the significance of German philosophy of language in relation to negative theology and the particular ways in which the inadequacy of language about God has theological consequences. Fourth, a more philosophical discussion of the ways in which what is necessary or possible in one language cannot adequately be conveyed, as necessary or merely possible, in translation.

This article is prompted in part by current debates at the end of the first quarter of the twenty-first century within the Church of England as to what pronouns can be used of God in liturgy.

It is now customary in Britain that the author of a journal contribution should identify clearly where the originality of a contribution lies to aid non-experts in evaluating the quality of its research. This contribution is arguably original for its cross-disciplinary synthesis of arguments from linguistics, medieval theology, feminist theology and German philosophy in order to present its argument. Its purpose is to extend the period of deliberation about the issue at hand (viz., what pronouns are liturgically permissible) by delaying the point at which a resolution of the issues is made. This is significant because of a tendency in much popular theology to name the desired conclusion and then to select only the reasoning supporting that conclusion. This piece resists attempts to reason back from a conclusion and presents a series of difficulties that obstruct a straightforward resolution. It has a conclusion, but its purpose is primarily to delay it in favour of deliberation.

Gender in English

God called the light Day, and the darkness he called Night. (Gen 1:5)

God has just separated light from darkness, and at this moment of the first separation God acquires a pronoun. ‘…the darkness he called Night’.

So God created humans in his image,

in the image of God he created them;

male and female he created them. (Gen 1:27).

God's creation of men and women in Gen 1:27 is the first we hear of sexes. God has already created many creatures including plants, fish, birds, land animals. Presumably they are sexed (we know the plants have ‘fruit with the seed in it’, according to v. 11) but it is not said. The first explicit appearance of sexual difference is in verse 27.

At the first appearance of sexual difference in scripture, God again acquires a gendered pronoun: ‘…in his image’.

Pronouns as a word class are a subject of much discussion by linguists, owing to the wide variety of uses in many languages, with attention drawn to the peculiarities of being a closed class of word but unlike others in that category, of being unusually indeterminate and contextual, with corresponding use of concepts like deixis and anaphora.¹ In a sentence: pronouns’ meanings radically depend on what is going on around them. Approaches that focus on intrinsic meaning (e.g. by inquiring into the definition of a word) fundamentally founder when it comes to pronouns. Whereas linguistic discussions are typically focused on questions of classification and change; however, the focus in what follows is on possibilities and consequences, especially in relation to translation. Translation is a focal concern in this discussion because the question, ‘What pronouns shall we use of God?’, is closely bound up with the question, ‘How shall we translate the Bible?’.

The citations above are from the New Revised Standard Version Updated Edition (NRSVue). When the NRSV was originally published in 1989, the editors made public their debate about gendered language. They recognised that in talk about men and women, the older convention of using ‘man’ for both was a problem. But they also acknowledged that their move towards inclusive language presented challenges because they were translating language whose origins were in a patriarchal culture:

During the almost half a century since the publication of the RSV, many in the churches have become sensitive to the danger of linguistic sexism arising from the inherent bias of the English language towards the masculine gender, a bias that in the case of the Bible has often restricted or obscured the meaning of the original text. The mandates from the Division specified that, in references to men and women, masculine-oriented language should be eliminated as far as this can be done without altering passages that reflect the historical situation of ancient patriarchal culture.²

Patriarchy names a social hierarchy with men at the top and women at the bottom. If you embrace and seek a patriarchal hierarchy, whatever your sex or your gender, the older conventions do not pose a translation problem. Patriarchal then? Patriarchal now. No problem.

But suppose you are translating a text from a patriarchal culture for use in a culture that rejects patriarchal hierarchy, or even pursues mobile hierarchies (I am your manager this year, but next year I am your assistant. Today I am President; tomorrow I am a farmer). What then?

An accurate translation, whether pursuing dynamic or formal equivalence, would preserve the patriarchy of the source.³ But it then becomes an obstacle to use. If one is part of a community that seeks to inhabit scripture but not inhabit patriarchy, there are hard decisions to make.

The NRSV team elected in the late 1980s to use inclusive language. For the most part this was restricted to its use for humans. It is not obvious what thought was given to pronouns for God. For the 2021 edition, the public discussion is even briefer.⁴ In both sets of editorial remarks it is noted that reading scripture in English introduces hazards that may not appear in other languages. These are compounded by the fact that many of those who speak English as a first language do not speak a second language. This is especially sharp in relation to scripture. If you only think in English, then your English translation of scripture is the only scripture you can imagine.

So what is this English-language gender bias?⁵ A comparison with German and French brings certain things to light.

In English: ‘I said to my dog, “sit” and she sat.’

In German: Ich sagte zu meinem Hund: ‘Sitz’, und er setzte sich.

In French: J'ai dit à mon chien « assis » et elle s'est assise.

In German the dog's actual sex is irrelevant. Female dog? Male pronoun, because a dog is er. In French a dog is le chien (masculine), but the pronoun can change if you know the dog's sex (although it doesn't have to). This is all well-known to multi-linguists.

I use ‘sex’ because the examples are those of pet animals. In discussions of gender and sex in humans it is common to use the terms ‘grammatical gender’ and ‘social gender’ to capture the relevant distinction.⁶ Grammatical and social gender are distinct, but not wholly separate.⁷ This is a topic of some complexity as evidenced by the UK Government's attempt (so far unsuccessful) to harmonise sex and gender categories for the purposes of data collection.⁸

Some interesting linguistic oddities result. Consider cats, who are masculine in French (le chat) and feminine in German (die Katze). For the most part, sentences with cats in French will include le chat and sentences in German include die Katze, regardless of what sex the cat actually is.

But in German you can say, Meine Katze wurde kastriert (‘My cat was castrated’). The feminine noun is quite normal. You can say mein Kater if you want to identify its sex as male, but you don't have to. (My thanks to native speaker Annedore Wilmes for this example.)

In French you can say, Mon chat est une chatte (‘My cat is a female cat’). Mon chat is masculine in this sentence. Again, this is normal usage. (My thanks to native speaker Isabelle Hamley for this example.)

The gendered nouns in German and French are in most cases unrelated to the sex of the dog or cat. The language furnishes possibilities for being specific, if it is relevant, but for the rest one uses the default gender. That means there is a good argument for saying that the default gender is unsexed. When one says le chat or die Katze one is not necessarily saying anything at all about its sex.

In English gendered language is more informative. If you say in German, ‘Ich sah die Katze an. Sie schaute mich auch an’ you are not identifying its sex. When we say, ‘The cat looked at me. She looked back’, we are in English, by contrast, identifying the sex of the cat.

Gendered language about God still matters in French and German. But arguably less. Every noun has to have a gender. So le dieu and der Gott. Strictly speaking this does not tell you the sex of God any more than le chat and die Katze tell you about the sex of cats. That is, the mere presence of a noun's gender is not intrinsically informative.

Of course, it may be that God's ‘social gender’ is in fact being specified by saying le or der, and then using the pronouns il or er. And it may be that God's social gender is being conveyed by the use of grammatically gendered terms. But this is something extrinsic to the grammatical use. More strongly, if this ‘something extrinsic’ were not added to it, to convey God's social gender, it would not be automatically conveyed by the grammatical gender. Certain assumptions can be carried by the grammatical gender, but the latter is a vehicle for them, not intrinsically an expression of them.

In English we choose a pronoun. We must. If we choose a male pronoun for a dog, or a female pronoun for a cat, we are determining the dog's or cat's sex, in a way that is not true in some other languages.

And if we choose a pronoun for God…? Whatever we are doing, what's going on in English is different from what goes on in French and German.

Let us consider the Latin translation of Genesis, in use for hundreds of years in Europe:

et vidit Deus lucem quod esset bona et divisit lucem ac tenebras (Gen 1:4)

et creavit Deus hominem ad imaginem suam ad imaginem Dei creavit illum masculum et feminam creavit eos (Gen 1:27)

Latin has pronouns, but typically they are not needed for verbs. The verbs divisit (v. 4) and creavit (v. 27) both lack them.

The Latin for ‘in his image’ is ad imaginem suam. Because ‘image’ is feminine in Latin, so God's image is, in a rather unexciting sense, feminine. It would be exciting in English, because English gets excited about sex in language. It's not exciting in Latin.

For hundreds of years, God's action in Genesis 1 was for Latin-reading western Christians ungendered. God acquired a gender in Genesis 1 in English only when translations first appeared. Ælfric's translation of Genesis was of Old Latin; only with Tyndale's translations was the Hebrew significant.⁹ Ælfric uses the pronoun hit (he) twice in the early verses of Genesis.¹⁰

Because English does not grammatically gender its nouns, as French, German, Hebrew, Greek and Latin do, any nouns that are gendered in English (e.g. mother, bullock – family and farming are prominent contexts for gendered nouns) or any verbs that acquire a pronoun (as they must) acquire a social gender and/or actual sex. If we want to avoid that for animals we often have to linguistically neuter them.

My opening proposal is that one should be aware of the difference between how pronouns function in English compared with other languages, even if one speaks no other languages.

Fatherhood: an incomplete figure

Christian speech about God is problematic. The question of which pronouns to use of God does not make it so. It is already problematic because speech is for things. And God is no thing.

There are, I propose, two relevant contradictions in play, in a consideration of Christian speech about God, and especially gendered speech. I want to explore these in dialogue with Janet Martin Soskice, who in turn is in dialogue with Paul Ricoeur.

The first contradiction is between longing for intimacy and struggling against idolatry. The second is between exegesis and theology.

Like as the hart desireth the waterbrooks,

So longeth my soul after thee, O God.

(Ps 42:1, Coverdale translation, modern spelling)

We long for God.

Here is a promise of intimacy, its possibility, its impossibility, its promise endlessly unfulfilled, memorably summarised by St Augustine in Confessions: our hearts are restless till they find their rest in thee. Coverdale's translation of Psalm 42 offers a pun on this theme (‘hart desireth’) that is obscured in the KJV (‘As the hart panteth’) and obliterated in the NRSV (‘As a deer longs’).

But the tradition is also haunted by the horror of idolatry.

They have hands, and handle not; feet have they, and walk not

neither speak they through their throat.

They that make them are like unto them

and so are all such as put their trust in them.

(Ps 115:7–8, Coverdale translation, modern spelling)

The longing for intimacy bends language towards likeness, towards human relationships, towards visibility, touchability. The fear of idolatry snaps language sharply back, the broken tablets, broken words, the prohibition on images. There is no smooth path between this longing and this fear.

The second contradiction, between exegesis and theology, in some ways mirrors the first. The exegete deals in representation. For Ricoeur there is in exegesis a virtue in attending to ‘the figures of God’, in all their concrete, evocative, raw materiality, with attention to whether God appears described in the third person or through the prophet's first person utterance, speaking for God.¹¹ By contrast, as Soskice notes, the theologian is conscious of the operation of metaphor, of the fact that all language about God is figurative.¹² What is brought close to us in the immediacy of exegesis is pushed away in the reflectivity of theology.

At the heart of these two contradictions – between intimacy and idolatry, between exegesis and theology – is God as father.

Those who like to count things have noted that whereas in the New Testament God is named ‘father’ between 170 and 190 times, and prayed to in this name, in the Old Testament there are not even a dozen such designations, none of them in prayer.¹³

One can discern a theological pattern: God's relation with Israel is a covenant with Abraham, and although this covenant is sometimes figured as kinship, especially as Ricoeur – ever attentive to genre – notes in the prophets, it is covenant and promise that order the narrative.¹⁴ When Moses asks for God's name in Exodus 3, it is ‘I am’: ungendered and unkinned. In this tradition God not paternal but is above all the one who makes a covenant with Abraham and who brings the people out of Egypt.¹⁵

In the New Testament something new happens at the start of Mark's Gospel and its synoptic parallels:

In those days Jesus came from Nazareth of Galilee and was baptized by John in the Jordan. And just as he was coming up out of the water, he saw the heavens torn apart and the Spirit descending like a dove upon him. And a voice came from the heavens, ‘You are my Son, the Beloved; with you I am well pleased.’ (Mark 1:9–11, NRSVue)

Jesus is named as ‘my Son’. God is thus implicitly identified, in relation to Jesus, as his father.

It is a moment of outrageous proximity, visibility, audibility, intimacy. It evokes and patterns prayer.

Here a third contradiction becomes visible: between undeniable patriarchy and undeniable trinitarian begottenness.

The patriarchy is undeniable. The traditions are slathered in it, from individual figures to entire dualistic systems of hierarchy. Soskice shines a bleak light on it in two sentences:

…what is objectionable is not simply that God is styled as male in the tradition, but that the ‘divine male’ is styled as one who is powerful, dominant, and implacable. This is disturbing not just in its subordination of women, but in giving divine justification to a hierarchical reading of the world conceived in the binaries of powerful/powerless, superior/inferior, active/passive, male/female.¹⁶

Whenever hierarchy is established, with some above and some below, the female is typically below with the male on top, typically responsive with the male taking initiative, typically following where the male leads, typically sensual where the male is rational.

As Soskice points out, the hard-line patriarchal theologians of today are not wrong, especially where they choose to be more literal minded: the Bible is a flood of patriarchal classifications in which women are drowned (my words, not hers). The subordination of women is endemic.

This subordination is endemic but not universal. The one place it does not extend is, surprisingly, fatherhood. God is patriarchally figured in the Old Testament, but rarely as father. And where God is figured as father in the New Testament, it is typically not patriarchal. God is father not because he is ‘like a father’, in the same way that God is a rock because he is ‘like a rock’. God is Father because Jesus is God's Son.

God is Father in two central cases: as one to whom one prays, and as the first person of the Trinity. Here something surprising happens to the paradoxes. God is Father in prayer, and thus in the longing for intimacy. But God is also Father in trinitarian doctrine, and thus in the struggle against idolatry. God is Father both in ‘unrefined’ exegesis and in ‘figurative’ theology, in both proximate address and distanced reflection.

So what is the meaning of ‘father’ here?

God draws near. Appallingly, dangerously, wonderfully, reassuringly near.

What marks Jesus’ speech about God, for Ricoeur, for Soskice, for us (?), is its audacity. To call God, ‘Father’!

God as Father is not some ancient symbol carved in stone, weighing down the tradition for centuries, but a new figure, supple, announced but as yet (in the New Testament) untested, unexplored, unknown. The Father is an incomplete figure, pregnant with possibilities.

Negative theology and German philosophy

The long Christian tradition has by and large agreed that God can be named with confidence, but only described with diffidence. By and large it has agreed that God can be freely addressed in prayer, but only spoken of, in the third person, with great care.

Creatures use creaturely language for creatures. So can it be used for what is not a creature, namely God? The long tradition has agreed that it can, but things will get strange.

Consider some famous philosophical puzzles. Leibniz famously pointed out that there is no concept for the colour blue. There is only its name. And you can only learn it by seeing it. No one can tell you what blue is. To use his terms, blue is a sensible quality, and thus lacks the nominal definition of intelligible (i.e. conceptual) qualities.¹⁷ Wittgenstein famously wondered why there is no concept for the smell of coffee. Do we lack words for it? Why would we think there should be words for it? You have to smell it for yourself.¹⁸ Hesiod famously wrote that he relied on the Muses to describe the gods, and this means receiving the ‘mythic’ forms where false things look true, and truths live in things made up.¹⁹ Gregory of Nyssa famously speaks of the luminous darkness.²⁰ One can have knowledge, as Moses does. One can approach God, as Moses does. But no one has knowledge of what God is, not even Moses, no matter the immensity of knowledge, no matter the nearness of approach.

Is it easier to describe God than the colour blue? Do we have surer words for God than for coffee? Is our self-reliance sufficient to bypass the Muses? Are we equipped to clarify the luminous darkness? These questions invite the answer, no. Yet we are not without experience, not without words, not without Muses, not without luminous darkness. In such cases, however, words are not adequate to things. And God is no thing.

Negative theology, or apophatic theology, or the via negativa, is not so much a way of doing theology, as it is a set of disciplines that one can learn after one has learned to speak of God in various ways.²¹

Hesiod, more than six centuries BCE, is instructive because he says not only, ‘This is the song’, but, ‘Muses give this song’. He draws attention to Muses. He says not only, ‘These are the words’, but, ‘These words are mythos’, which we might translate as ‘word’ or ‘wording’. He draws attention to the wording, which later thinkers will translate more directly as myth. Gregory of Nyssa, in the fourth century, is instructive because he highlights the strange things that happen to language when it approaches God – when it does what Moses did. It produces artefacts like ‘luminous darkness’. Leibniz in the late seventeenth century and Wittgenstein in the middle of the twentieth century are instructive because they draw attention to names that do not, cannot, communicate the describable properties of what they name.

Negative theology is a secondary discipline, learned after the primary discipline of speech.

One learns to say ‘blue’ and ‘smell of coffee’ long before one learns to say the kinds of thing Leibniz and Wittgenstein say. One learns to say ‘God’ and ‘scripture’ long before one learns to say, ‘luminous darkness’ and mythos.

The secondary discipline of negative theology is simultaneously ancient, in Hesiod and Gregory, and advanced intellectual technology. It is the fruit of study. The secondary discipline of negative theology is simultaneously everyday, in Leibniz and Wittgenstein, and sophisticated philosophical insight. It is the fruit of perplexity.

Negative theology explores parallel issues:

Is God large?

Theology: God is enormous.

Negative theology: God is beyond large and small, beyond size.

Is God strong?

Theology: God is stronger than anything.

Negative theology: God is beyond strength and weakness, beyond measure.

Is God visible or invisible?

Theology: God is visible. God is invisible.

Negative theology: God is luminous darkness, dazzling darkness, beyond visibility and invisibility.

God is ‘beyond’ the divisions. The common word for this is transcendence.

The traditions of German philosophy, especially in the period between the French Revolution and the defeat of Napoleon, make these issues of transcendence a focus of intense study. The German traditions both inherit and expand the traditions of negative theology. They inherit forms of thought that stress the failure of language to grasp its object. They expand these forms to embrace objects other than God: the ground of articulation; the whole of which things are parts; the absolute origin of consciousness, subjectivity, mind. In the traditions of negative theology God is no thing. The traditions of German philosophy expand the range of not-things.

German philosophy receives the traditions of negative theology, incubates its own distinctive insights and then becomes available to theologies which can in turn receive traditions that have become alien to it.

The most famous lessons can be summarised briefly: the ground of articulation cannot be articulated; the genesis of consciousness is not something we can be conscious of; the condition for meaning is not something that can be meaningfully grasped; and so forth.

The role of language has several dimensions in German philosophy, as Andrew Bowie has argued in his brief but sophisticated Very Short Introduction to German Philosophy, in the chapter on ‘The Linguistic Turn’.²² To summarise in one brief sentence: we are dependent on language. This is a significant and challenging idea in a period when human autonomy, freedom, independence are major themes in the wake of the American and French Revolutions.

Language connects and alienates us. It connects us when we share a language. It alienates us when we don't. It is thus linked to tribalisms of various kinds, including nationalisms in a period of European wars.²³

Language both represents the world as if humans don't exist, an aspiration to objectivity vis à vis what is ‘already’ there, and expresses ‘our’ world and our meaning in it, conveying our subjectivity vis à vis what is unfolding.²⁴

To reflect on language is to discover that what is objective is already subjective: we use language to describe things; and what is subjective is already objective: we use language to describe things, and that language is already in play. We cannot step outside language to compare it to the world. The world is already ‘in’ language, and language is already ‘in’ the world.

All reasoning is shaped by language; there is no universal language, so there is no universal reasoning. There are multiple reasonings in multiple languages that are connected via fragile but generative acts of translation. Even the purest scientific experiment, which might appear a candidate for a universal language, relies on a prior linguistic orientation to the world, and that orientation, because it is linguistic, is shaped by a particular language and the particular histories it embodies.

Language is a system of signs, in which signs point to things and other signs. But what connects signs and things is not intrinsic to the signs and things themselves. The connection is established by use. And use is another name for particular histories. It is inescapably contingent.

Language is a changing system of signs, where the web of interconnected meanings is under constant renegotiation as cultures engage and are shaped by each other.

In sum, anything that appears in language will be shaped by historical, cultural, objective, subjective, rational, contingent, free, spontaneous, determined and rule-bound dimensions. And everything appears in language, including scripture.

The long tradition of theology had a primary practice of speaking of God, with confidence and a certain boldness. It also had a secondary discipline of negative theology which explored in various ways the precariousness of language about God. The primary practice is embodied in exegesis (Ricoeur again), liturgy, prayer and preaching. The secondary discipline is embodied in theology, in philosophy, in mystical practices of contemplation.

Such practices are unevenly distributed in social life. For most people, most of the time, quite respectably, one is engaged in the primary practice. It is a nice question when the secondary disciplines come into play. Perhaps they do so when the primary practices are under pressure, or even threatened.

As well as the insight that what is no thing cannot be grasped by language, and especially the species of no thing that functions as a ground, or origin, or whole, the traditions of German philosophy contribute something else of significant use for theology: a shift away from the idea that language represents states of affairs to a view of language as expressing relations, including our relations to the world.

Language here is not primarily a medium in which something that exists can be coded, transmitted, stored and decoded, in a way that leaves the ‘something that exists’ untouched or merely recorded. It is more than a medium. It is a way of doing something, a way of enabling something to be done, even itself something that has effects.

Whereas in negative theology language is both necessary and unstable when it approaches God, in German philosophy language is necessary and unstable as such.

Whereas in negative theology there is deep meditation on the ways in which language expresses our relation to God, and the possibility of our transformation in that relation,²⁵ in German philosophy language expresses our relation to everything.

Our Christian theological traditions produced German philosophies, and now receive them back as prodigal son. The embrace is rather uncertain. On the one hand English-language philosophy is dominated by analytic philosophy, which has an uneasy although ever-changing relation to the German tradition. At the same time, most theological discussion is guided by its primary practices, and the German tradition is a reflection and development of its secondary disciplines. Multiple negotiations are in play.

The traditions of negative theology have tended to recognise God's transcendence through linguistic experiments, including paradoxical formulations. Pronouns are often considered a closed class of words, that is, in most languages one makes do with existing pronouns rather than adding new ones. One of the interesting linguistic features of negative theological formulations is that, as in the case of pronouns, they tend not to introduce new terms, nor to change the meanings of old terms, but experiment with unusual juxtapositions, whether of words (such as ‘luminous darkness’) or of contexts (such as the approach to God by Moses in Exodus 19) – or of both at the same time.

The question, ‘What are God's pronouns?’, invites further questions about whether and to what extent language grasps God. Pronouns are radically dependent on context for their meaning. Experimenting with words and context, especially at the same time, is likely to have unpredictable consequences for pronouns, bringing to light new things about old habits, and making new things, new habits, possible.

‘He called Night’

Some philosophers in the twentieth century and beyond favour a distinction between referring to things (‘that one’) and meaning things (‘that one is poisonous’). This is a distinction made by Gottlob Frege in the 1890s. It is now a philosophical commonplace, although there remains lots of argument about how precisely this distinction works.

Very roughly:

You can point to something: that is a reference.

What is meant by something: that is a meaning (or sense).

As a consequence, you can point to something but not know its meaning: ‘What is that?’ And you can know something's meaning but not know its reference: ‘Where can I find mushrooms?’

The question, ‘Is human language adequate for God?’, thus has within this little system of classification two possibilities that can be expressed as questions: ‘Can we refer to God?’, and ‘What is meant by God?’

The long Christian tradition, in dialogue with other traditions, has experimented with various claims in relation to reference and sense. The majority view is: we can point to God adequately, but saying what is meant by God falls short.

That does not mean we cannot describe God. It means that our descriptions are limited and inadequate in various ways. The primary function of creaturely language is to point to (reference) and describe (meaning or sense) other creatures. When Adam first uses language in the garden, it is in reference to other creatures: ‘The man gave names to all cattle and to the birds of the air and to every animal of the field’ (Gen 2:20). (I wonder if perhaps he also, at the same time, named imaginary animals like the unicorn and pegasus, because that's what humans do.)

But Adam does not name God.

The first ‘sense’ of God, within the drama itself, is the serpent's, ‘Did God say..?’ And the second is the woman's as she echoes it: ‘…but God said…’.

God is first named, in speech, by the serpent.

Eve, and then presumably Adam, and then presumably all of humanity, learn to name God from the serpent.

The serpent is the first theologian. The serpent makes claims about what God knows: ‘God knows that when you eat of it your eyes will be opened, and you will be like God, knowing good and evil’ (Gen 3:5). The first theologian is, interestingly, not an exegete but a philosophical theologian.

Eve is the second theologian. She names God as the one who speaks a command: ‘… but God said, “You shall not eat of the fruit of the tree that is in the middle of the garden, nor shall you touch it, or you shall die.”’ (Gen 3:3)

The first theology is not addressed to God, in the second person. It is speech about God, in the third person, when the serpent and the woman converse about the fruit.

Adam is the third. He never names God. But he reasons, in a rather vague way, about God's possible actions. ‘I heard the sound of you in the garden, and I was afraid, because I was naked, and I hid myself’ (Gen 3:10).

Adam does not name God, but he is the first person to address God as ‘you’. If all speech to God is prayer, then this is the first prayer. And the first prayer is a declaration of fear.

The order of events in Gen is thus: philosophical theology first; prayer second.

God acts for the most part. There are no descriptions of God, other than descriptions of God's actions. God's ‘sense’ is action, not predicates.

It is not until there are cities that anyone names God. We are not told what they name God. We are just told, ‘At that time people began to invoke the name of the Lord’ (Gen 4:26).

The first description of God comes quite late: ‘And the Lord was sorry that he had made humans on the earth, and it grieved him to his heart’ (Gen 6:6). The first predicate made of God is that of disappointment. It, too, is expressed through a verb, an action: nacham YHWH (‘he-repented God’).²⁶

It is perhaps not surprising that theologians who are attentive to scripture think of God's acts when they think of the ‘sense’ of God. God appears more as one who acts, and less as one who is described.

So what about God's pronouns?

In Greek and Latin, verbs are not gendered. In Genesis 1, when God creates, the verbs epoiēsen and creavit are not gendered. The nouns theos and deus are masculine nouns. So God is grammatically gendered nominally but not verbally.

In English, neither nouns nor verbs are gendered. ‘God created.’ God is neither nominally nor verbally gendered.

When a verb is used in Greek and Latin, it can be used by itself. Epoiēsen and creavit are perfectly adequate: ‘x created’, where x is unspecified. The sentence ‘In principio creavit caelum et terram’ is complete, even if the noun is removed. It says that someone, as yet unspecified, created the heavens and the earth. Unspecified, ungendered.

It is not possible to translate this into English directly. In English if a verb is used without a noun, a pronoun must be substituted. We cannot properly say, ‘Created the heaven and the earth.’ We have to say ‘he created’, or ‘she created’ or ‘it created’ or ‘they created’ (the singular forms available to us).

Hebrew is different yet again. Elohim creates. Using a grammatical analytical framework suitable for Latin, one can say that elohim is masculine plural. But what would the masculine singular be? It does not appear in the text, or any text, and those who discuss these things find themselves looking at Ugaritic parallels.

Perhaps it would be wiser to use categories from medieval Hebrew grammars. Their concern is in part to compare Hebrew with Arabic, not Latin.²⁷ We could inquire as to what they say about plural nouns like Elohim, and we might find that they are attentive to the status of the one referred to (like the ‘royal we’ in English). But such grammars, from the tenth century, are themselves late, while the text in question is early.

In English (and Greek and Latin and many other languages) a plural noun takes a plural verb.²⁸ English verbal forms have an unusual morphology. I create. You create. He/she/it creates. We create. You (pl.) create. They create. We only have two distinct forms: create and creates. They are not neatly distributed between singular and plural. The form ‘creates’ is only used for he/she/it. If the singular form ‘they’ is used, it takes the same form as the plural: ‘Someone left their hat; I wonder who they are.’

This is not so in Hebrew. The plural noun Elohim of Genesis 1:1 takes a singular verb bara: he-created (the hyphen, again, denoting a masculine verbal form).

The first time God is gendered in Genesis is not in the (plural) masculine noun Elohim, but in the (singular) masculine verb bara, which precedes it.

Firstly (bereshit) he-created (bara, singular) God (Elohim, plural) the heavens and the earth.

Incidentally, heavens is plural in Hebrew (hashamayim) and singular in Greek (ton ouranon) and Latin (caelum), and in the King James English (‘the heaven’), but plural again in the NRSV (‘the heavens’).

There are no pronouns as yet. There won't be any pronouns in the Hebrew for a while because they are not needed. The verb form contains it, as in bara (‘he-created’).

As noted earlier, in English a pronoun appears early on, in verse 5: ‘God called the light Day, and the darkness he called Night.’ But ‘he’ is not in the Hebrew (as a pronoun), or the Greek or the Latin: qara (‘he-called’), ekalēse (‘x-called’), appellavit (‘x-called’). The verbs require no pronouns. In order to translate the Hebrew masculine verb form an English pronoun seems to be needed.

It is questionable whether that masculine verb form needs to be translated in this case, however: it is, after all, not translated in the Latin and Greek. Hebrew verb forms are necessarily gendered, and in Hebrew the only options are masculine and feminine (unlike Latin and Greek, Hebrew has no neuter gender). English has the possibility for gendering a verb by adding a pronoun. What must be gendered in Hebrew can be gendered in English by adding a pronoun: in the act of translation, the necessity of the gendered verb form does not travel from Hebrew to English. What is necessary in Hebrew is merely possible in English. In translation the necessity becomes possibility. This is significant. To say something because you have to is different from saying something because you choose to. A Hebrew verb ‘has to’ gender any verb attached to God. By contrast, in English, one chooses to attach a gender to God in relation to a verbal form.

It was previously noted that in French and German the grammatical gender (le chat, die Katze) does not necessarily indicate the sex of the animal. In English, to use ‘he’ or ‘she’ of an animal is not to indicate its grammatical gender but its sex. By extension, use of the grammatical gender of God in French and German (le dieu, der Gott) does not necessarily indicate the sex of God. In English, because there is no such thing as grammatical gender, to use pronouns is to indicate God's sex. It was noted that it is possible in French and German to indicate sex (la chatte, der Kater). In the light of the Hebrew discussion, we can add that, by contrast, in English indicating sex is necessary consequence of using a pronoun. What is merely possible in French and German is necessary in English.

The following odd feature of translation thus becomes visible: when translating gendered verb forms (Hebrew) or grammatically gendered nouns (French and German) into English, one is not just specifying or not specifying sex: one is engaged in modal transformation from necessity to possibility (Hebrew to English) and possibility to necessity (French/German to English). What is not – and cannot be – preserved is the necessity and the possibility: these are irreversibly changed.

A brief analogy may throw this into sharper relief. If you do something out of necessity, and I choose to imitate it, I am not imitating the necessity; I am imitating by choice the thing done necessarily. This is for example what happens when a non-disabled person mocks a person with a disability by imitating them: they mimic the action not the disability (because it is not, for them, a disability). Likewise, if Hebrew does something out of necessity, and a translator chooses to render it in English, the translation does not translate the necessity, only the thing that was done necessarily; but now it is done out of choice.

It is perfectly possible not to translate the gendered verb form into English. Verse 5 could (following the Latin) be translated like this: ‘God called the light Day and the darkness Night’ (Appellavitque lucem Diem, et tenebras Noctem). Luther follows this in his various German translations between 1534 and his death: …und nannte das Licht Tag und die Finsternis Nacht.

Perhaps because the verb is repeated, in both the Hebrew and the Greek texts, modern translators repeat it in English, thus choosing to add a pronoun which is not there in Hebrew, Greek or Latin. It is a choice, a matter of possibility rather than necessity. They could repeat the noun: ‘God called the light Day, and the darkness God called Night.’

What is God's pronoun? Very often it is English that chooses, and thus forces, a pronoun where it is not necessary in the ancient languages, because other necessities supervene.

Like English, verbs in German and French need pronouns where the noun is not specified. Unlike English, in French and German all nouns are gendered and the pronoun used is often determined by the noun's grammatical gender, as in the example of dogs and cats: the gender of the noun tells you nothing about the sex of the animal referred to. For social terms like Mann/homme or Frau/femme it is admittedly different: grammatical gender expresses social gender. Furthermore (as noted above), in German, social genders are invariably assigned for professions: a doctor is either der Arzt or die Ärztin. In French the opposite happens, and gender can be lost in translation: ‘The doctor will see you in her office’ is Le médecin vous reçoit dans son cabinet. But ‘God’ is not a profession.

In Hebrew pronouns are needed less often than they are in English because, unlike the other languages considered here, Hebrew verbs are gendered.

What is a ‘biblical’ answer to this question about God's pronouns? It depends whether the Bible is in Hebrew, in Greek or in Latin.

Theology is an unusual discipline for granting greater authority to older sources. One can thus ask: which text is older?

These can be arranged in order of age. There are some Hebrew manuscripts among the Dead Sea Scrolls from 150 BCE to 75 CE, and these show that there was no single version of the Hebrew text. The earliest manuscripts for the Septuagint (Greek) text are from the second century BCE. The earliest Vetus Latina manuscripts (a Latin version that pre-dates Jerome's translation) are from 450. The earliest Vulgate manuscripts of the early books of the Bible are Pentateuchus Turonensis from sixth–seventh century. The earliest manuscripts for the Masoretic text, which is the authoritative Hebrew version for Rabbinical Judaism and for Christian translators, are from the ninth century CE.

What about translations into English?

The earliest of these were translations of the Vulgate, whose first complete manuscript, Codex Amiatinus, dates from the eighth century. The earliest translations to use Hebrew and Greek were by Tyndale between 1522 and 1535.

So which of these is ‘biblical’? They all are: where there are manuscripts, there we find texts that are authoritative for a community.

We only have manuscripts. Were we to consult a rabbi, perhaps she would say, ‘The only texts written by God were not biblical; they were the two tablets of the covenant brought down by Moses, and broken. After that, everything is written by humans.’

What did God write on those soon-to-be-broken tablets? It is not specified. ‘When God finished speaking with Moses on Mount Sinai, he gave him the two tablets of the covenant, tablets of stone, written with the finger of God’ (Exod 31:18).

People have speculated that these must have been the Ten Commandments. They are often called in Christian paintings the Tablets of the Law, although in Exodus they are the tablets of the covenant (in Hebrew) or the tablets of the testimony (in Greek and Latin).

Were they written in Hebrew? If so, perhaps it was Hebrew from the thirteenth century BCE, the time of Moses. What kind of Hebrew was that? We do not know. The oldest Hebrew we have are inscriptions in so-called ‘Biblical Hebrew’ from the tenth century BCE. Most are a single letter or word. It has been argued that the very idea of a singular ‘Biblical Hebrew’ is misleading, because many linguistic practices are conveyed by this name. It is, for example, not possible to distinguish ‘correct’ from ‘errant’ grammar, or ‘literary’ from ‘vernacular’ uses when one has only written texts and no native speakers to consult.²⁹

It is often considered important by some Christians to determine whether one accepts a literal interpretation of scripture. But a literal interpretation of what? Perhaps of the earliest manuscripts, which are not used today by any community. Perhaps of later harmonisations of manuscripts, accepted as the ‘received text’ by large numbers of communities today.

However we approach these matters of trace and transmission, the question of what pronouns to use is accompanied by the question of what language to use those pronouns in, and indeed whether pronouns are needed at all.

The question thus ceases to be, ‘What pronouns for God shall we use?’ A prior question arises: ‘Who is using pronouns, and in which language are they using them?’ More directly: ‘What pronouns for God shall we use in English?’

Answering that question may involve asking what pronouns are used in other languages, upon which English is dependent, and with which English is in constant dialogue and negotiation.

Is it true that God's verbs are masculine in Hebrew? Yes. Is it true that God's nouns are masculine in Hebrew, Greek and Latin? Yes for the most part.

Is it true that God the Father and God the Son are masculine nouns, while God the Spirit is feminine? No. Ruach (Hebrew) is feminine. Pneuma (Greek) is neuter. Spiritus (Latin), Geist (German), Esprit (French) are masculine. Both ‘Ghost’ and ‘Spirit’ (English) are ungendered, because all nouns are in English ungendered.

Even in languages with grammatical gendered nouns, can you infer something's sex from the gender of its noun? Not always directly and sometimes not at all. Germans do not think that all cats are female, and French people do not think that all doctors are male. It is not obvious that French people think that all gods are male: this cannot be inferred from the grammatically gendered le dieu/les dieux. (And most theologians trained in the long tradition, French or not, would in any case deny that God is ‘a’ god.)

In English something distinctive happens. God is not grammatically gendered. Nothing is. So when God's pronoun appears, as it might in Gen 1:5, this does not follow a grammatical gender. There is no grammatical gender to follow. It performs a quite new function: it specifies God's social gender. In this respect English is unlike many other languages.

Asking about God's pronouns in English is different from asking about them in other languages. In other languages, to specify a pronoun is often to repeat the grammatical gender. But in English, to specify a pronoun, where the noun is indeterminate, is typically to specify a social gender.

Is God figured as male in patriarchal societies? Yes, for the most part, but not always. There is also feminine imagery not just for God as creator, but also for Jesus. And not just masculine and feminine. The more scripture and tradition one considers, the more interesting this question is. Consider the following:

Jesus as self-wounding pelican in the Physiologus.

Christ as mother in Julian of Norwich.

God as light in Isaiah 60.

God as deep water, as fountain, in John of the Cross.

Holy Spirit as God's kiss in Bernard of Clairvaux.

God as rock that fathers, that gives birth in Deutoronomy 32.

What follows from all this? No single thing follows from it.

Language is strange. It gets especially strange as it approaches God.

There is no language, only languages. There is no divine pronoun, only pronouns.