0% found this document useful (0 votes)
173 views80 pages

Timothy Pawl - The Incarnation (Cambridge Elements)

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
173 views80 pages

Timothy Pawl - The Incarnation (Cambridge Elements)

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 80

PAWL

The doctrine of the Incarnation, that Jesus Christ was both


truly God and truly human, is the foundation and cornerstone
of traditional Christian theism. And yet this traditional teaching
appears to verge on incoherence. How can one person be
both God, having all the perfections of divinity, and human,
having all the limitations of humanity? This is the fundamental Philosophy of Religion
philosophical problem of the Incarnation. Perhaps a solution
is found in an analysis of what the traditional teaching meant
by person, divinity, and humanity, or in understanding how
divinity and humanity were united in a single person. This
Element presents that traditional teaching, then returns to the
incoherence problem to showcase various solutions offered

The Incarnation

The Incarnation
to it.

About the Series Series Editor


This Cambridge Elements series provides Yujin Nagasawa
concise and structured introductions to University of

Timothy J. Pawl
all the central topics in the philosophy of Birmingham
religion. It offers balanced, comprehensive
coverage of multiple perspectives in the
philosophy of religion. Contributors to the
series are cutting-edge researchers who
approach central issues in the philosophy
of religion. Each provides a reliable
resource for academic readers and
develops new ideas and arguments from a
unique viewpoint.

Cover image: Eerik/ iStock / Getty Images Plus Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to ISSN 2399-5165Core
the Cambridge (online)
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341 ISSN 2515-9763 (print)
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
Elements in the Philosophy of Religion
edited by
Yujin Nagasawa
University of Birmingham

THE INCARNATION

Timothy J. Pawl
University of St. Thomas

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre,
New Delhi – 110025, India
79 Anson Road, #06–04/06, Singapore 079906

Cambridge University Press is part of the University of Cambridge.


It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781108457521
DOI: 10.1017/9781108558341
© Timothy J. Pawl 2020
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2020
Nihil Obstat Imprimatur
Reverend George Welzbacher The Most Reverend Bernard Hebda
Censor Librorum Archbishop of Saint Paul and Minneapolis
November 5, 2019 November 5, 2019

The Nihil Obstat and Imprimatur are official declarations that a book is free of doctrinal
error. No implication is contained therein that those who have granted the Nihil
Obstat and Imprimatur agree with the content, opinions, or statements expressed. Nor
do they assume any legal responsibility associated with publication.

A catalogue record for this publication is available from the British Library.
ISBN 978-1-108-45752-1 Paperback
ISSN 2399-5165 (online)
ISSN 2515-9763 (print)
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation

Elements in the Philosophy of Religion

DOI: 10.1017/9781108558341
First published online: September 2020

Timothy J. Pawl
University of St. Thomas
Author for correspondence: Timothy J. Pawl, timpawl@stthomas.edu

Abstract: The doctrine of the Incarnation, that Jesus Christ was both truly
God and truly human, is the foundation and cornerstone of traditional
Christian theism. And yet this traditional teaching appears to verge on
incoherence. How can one person be both God, having all the
perfections of divinity, and human, having all the limitations of
humanity? This is the fundamental philosophical problem of the
Incarnation. Perhaps a solution is found in an analysis of what the
traditional teaching meant by person, divinity, and humanity, or in
understanding how divinity and humanity were united in a single
person. This Element presents that traditional teaching, then returns to
the incoherence problem to showcase various solutions offered to it.

Keywords: Incarnation, Conciliar Christology, theism, metaphysics

© Timothy J. Pawl 2020


ISBNs: 9781108457521 (PB), 9781108558341 (OC)
ISSNs: 2399–5165 (online), 2515–9763 (print)

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
Contents

1 Introduction 1

2 The Person 5

3 The Divinity of Christ 10

4 The Humanity of Christ 13

5 The Hypostatic Union 23

6 The Activities of Christ 27

7 The Fundamental Philosophical Problem 36

8 Conclusion 54

References 55

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 1

1 Introduction
The doctrine of the Incarnation is the teaching that Jesus Christ, the human
crucified under Pontius Pilate, was truly God, one person of the Blessed Trinity.
That doctrine, affirmed by the orthodox statements of faith from the early church
through the later Catholic, Orthodox, and Protestant doctrinal statements, teaches
that this person, while truly God, became truly human, a human like other humans
in all ways except sin, for the sake of saving humans from their sins and bringing
them to perfect union with God. This doctrine is a fundamental part of any
traditional Christian teaching – if it is false, so is traditional Christianity.
While we may have become accustomed to it, the doctrine of the Incarnation
is a shocking claim. How could God become human? How could one person be
two seemingly incompatible types of things at the same time? If God is
conceptualized in the traditional sense as immutable, impassible, and simple,
how could such a person become changeable, affectable, and complex, as all
humans are? And why? Why would God do such a thing as to become human?

1.1 Methodology
The Incarnation has been analyzed from multiple perspectives. For instance, do
the Christian Scriptures imply that Jesus is true God, one in being with the
Father?1 Does the historical record give justification for belief in the existence
of the flesh-and-blood human named “Jesus,” son of Mary? Do the early
theologians affirm the doctrine? And so on. All of these scriptural and historical
questions are worthy of analysis. This short Element, situated as it is in a series
on the philosophy of religion, does not address these questions. Rather, it
focuses on the philosophical questions surrounding the doctrine. Such philo-
sophical analysis no doubt assumes some findings of these other methods of
assessment. For instance, it assumes, for the sake of argument, that the human,
Jesus, did exist.2 But it does not and could not, given length constraints,
responsibly enter into the other discussions.
While this Element focuses exclusively on the philosophical questions con-
cerning the Incarnation, its work cannot be done in a historical vacuum. The
doctrine of the Incarnation was formulated in a series of ecumenical councils –
the context of the statements of those councils must be taken into account to
understand the meaning they had for those speakers. This Element makes

1
The biblical justification for the doctrine of the Incarnation is vitally important. This Element is
not the place to enter into that vast literature. It is impossible to provide a brief set of paradigmatic
references to the ocean of literature on the biblical case for the doctrine of the Incarnation.
I suggest the reader start with Bird et al. (2014), Loke (2019), and Tilling (2015) and follow the
footnotes into the wider literature.
2
For scholarship on the historical case for the existence of Jesus, see Ehrman (2013).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
2 Philosophy of Religion

frequent reference to the doctrinal statements of these councils concerning the


Incarnation.
Many of the philosophers who have considered the philosophical questions
concerning the Incarnation have done so from a specific heritage. For instance,
many who believe in the Incarnation would be reluctant to affirm something that
contradicts the teachings of the ecumenical Council of Chalcedon in AD 451.
Likewise, many who object to the Incarnation want their objections to target the
traditional teaching and would see it as a misfire if their objections showed that
a view of the Incarnation not held by anyone in Christian history is incoherent.
For this reason, this Element notes whether a certain line of philosophical
thought is precluded by the orthodox statements concerning the Incarnation.
Throughout I make use of the excellent translation edited by Norman Tanner
(1990), Decrees of the Ecumenical Councils. In the hopes of providing some
evidence that I have not interpreted these texts in a nontraditional manner, I have
sought out and received a Nihil Obstat and Imprimatur on this Element from my
local Catholic diocese, the Diocese of Saint Paul and Minneapolis.
A note on language: the word “orthodox” is often used as a term of praise,
and its contrary, “nonorthodox” or “heretical,” is often used as a term of
derision. In this work, “orthodox” is stripped of such resonance, and I make
no use of “heretical” at all. (Here I’ve mentioned it twice.3) As I use the term,
orthodox teaching is that which is endorsed in the earliest seven ecumenical
councils, and a proposition is “nonorthodox” when it is proscribed by those
same councils. The reader will see that these definitions leave a large middle
ground: things neither taught nor proscribed by those councils, for instance,
the claim that Christ knew English. There is debate about what ought to count
as orthodox. I have no room here to enter into that debate. I stipulate instead
a meaning that I think finds consensus throughout the history of Christian
thought.
This Element presents the Christology promulgated by the first seven ecu-
menical councils of Christianity: the First Council of Nicaea in 325, the First
Council of Constantinople in 381, the Council of Ephesus in 431, the Council of
Chalcedon in 451, the Second Council of Constantinople in 553, the Third
Council of Constantinople in 680–681, and the Second Council of Nicaea in
787. I refer to the Christology of these councils as Conciliar Christology.
A final point about method: this Element speaks in a factive voice, as if the
Incarnation in fact occurred. Appending “according to Conciliar Christology”
or the like to each sentence would be tiresome for both author and reader. The
reader who does not affirm the Incarnation and who finds such factive

3
For helpful reflection on the concepts of “orthodoxy” and “heresy,” see Stump (1999).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 3

presentation unpalatable ought to treat the remainder of this Element as resting


on an assumption made for the sake of argument. Assume that the world is as the
Christians have believed, and that there was an Incarnation. In fact, we don’t
even need so strong an assumption. We can simply assume the possible truth of
the doctrine and still let this Element do its work.

1.2 What the Doctrine of the Incarnation Says


The teachings of Conciliar Christology can be organized under five topical
headings: the person of Christ, the divine nature of Christ, the human nature of
Christ, the union between the two natures (i.e., the hypostatic union), and the
activities of the God-man, Jesus Christ. In this Introduction, I offer an initial
comment about conciliar teaching on these five topics.
To the first point: Jesus Christ is one person, the Second Person of the Trinity,
referred to also as “the Son” and “the Word.” There are not two persons in the
Incarnation, one divine and the other human. Rather, there is one person who is
both truly divine and truly human because this one person had and has two
distinct natures.
To the second point: one of those natures was the one and only divine nature,
which Jesus possessed with and no less than the Father and the Holy Spirit. God
the Son made human, Jesus, was immutable and impassible in virtue of the
divine nature. His incarnation in no sense diminished his possession of that
divine nature.
To the third point: the other nature was a human nature that the Son took up
into himself in the Incarnation (the technical term is assumed). In virtue of this
human nature Christ was like us in all ways, sin alone excluded.
To the fourth point: these two natures were united in a unique manner. This
union is traditionally referred to as the hypostatic union. It is hypostatic, insofar
as the union occurs in the person of the Son, and person, as we see in Section
2.2, is a species of the genus, hypostasis. It is this hypostatic union between the
natures that allows for the Communication of Idioms (Communicatio
Idiomatum), the true description of the one person using a subject term drawn
from one nature and a predicate term drawn from another. To give a common
example, Paul says in 1 Corinthians 2:8 that the leaders of the world crucified
the Lord of glory. Here the “Lord of glory” (a term correct of Christ in virtue of
his divinity) is “crucified” (a term correct of Christ in virtue of his humanity).
To the fifth point: Christ died, rose from the dead, and ascended into heaven.
He willed. He knew. Many questions arise from such observations. How can it
be that God dies? Did he will evil? Could he have willed evil? Could he have
been tempted? Could he have been mistaken?

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
4 Philosophy of Religion

1.3 The Main Metaphysical Problem


Predicating terms of one and the same person drawn from two different natures
leads to difficulties, as the councils themselves make evident. The traditional
documents of Christianity assert both classical theistic attributes of Christ (e.g.,
impassible, immutable) and also mundane human attributes (e.g., suffered,
changed). Sometimes, as we see in Section 7.1, they predicate both apparently
contradictory predicates of Christ in the very same sentence!
It is these apparent contradictions in attributes that lead to what has been
called the Fundamental Philosophical Problem of the Incarnation (see Cross
2011, 453). How could one person be both God and human? For to be truly God,
that person would have to be eternal, omnipotent, omniscient, immutable,
impassible, and all the rest. But to be truly human, that person would have to
be temporal, weak in some ways, ignorant of some facts, able to change, able to
be affected, and so on. And nothing can be both eternal and temporal, or
omnipotent and weak, or . . . And so nothing can be both God and human.
Thus, Christ is not both God and human, and so the doctrine of the Incarnation is
false.
The Fundamental Philosophical Problem is the most important philosophical
objection to the doctrine of the Incarnation. But to answer an objection to
a doctrine properly, one must know what the doctrine says, which is why this
objection comes at the end of this Element, after explaining the doctrine.4

1.4 The Shape of Things to Come


This Element is composed of eight sections. The Introduction, now concluding,
discusses the methods I employ and offers some caveats. For instance, the
method is analytic, with attention paid to the traditional formulations of the
doctrine. The caveats include the claim that this Element does not assess biblical
or historical evidence for or against the doctrine.
Section 2, on the person of Christ, addresses the definitions of the key terms,
“hypostasis” and “person,” the names of the person (e.g., Logos, Son), and the
divine status of the person. Section 3, on the divinity of Christ, discusses the
divine nature and its attributes from two perspectives: the traditional view and
the kenotic view. Section 4, on the humanity of Christ, considers human nature
itself, whether it is concrete or abstract, whether it is composed of parts and if so
how many, whether it has a will and a mind, whether it is a person in its own

4
This Element focuses on the metaphysics of the Incarnation. Those interested in the atoning work
of the Incarnation should see Stump (2019) and, in this same Elements series, Craig (2018). For
discussion of epistemic questions concerning the acceptance of doctrine, including the
Incarnation, see McNabb (2018) from this same Elements series.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 5

right, whether it is fallen or unfallen, and the weaknesses or infirmities it has.


Section 5, on the hypostatic union, discusses whether the union can be fully
analyzed philosophically, what features it has of itself (e.g., is it created?), and
what we can say of the Communication of Idioms from the natures to the person,
in light of the union. Section 6 focuses on the activities of Christ, including his
will (Was it free? Could he sin?), his knowledge (Was he mistaken?), his death,
and his descent into hell. Section 7 discusses the main philosophical objection,
which is that anything divine must have certain predications true of it, but
nothing human can have those predications true of it; thus, Christ couldn’t be
both divine and human, contrary to the orthodox doctrine. The last section is
both more complex and longer than the previous sections, insofar as it takes up
many of the extant responses to the problem, some of which are quite technical.

2 The Person
The traditional doctrine of the Trinity teaches both that there is only one God
and that there are three divine persons: the Father, the Son, and the Holy Spirit.
Only one of these three persons, the Son, became incarnate. This Element
focuses exclusively on the Second Person of the Trinity, the one who, according
to Scripture and council, became human.
The focus on this person should begin at the outset with a discussion con-
cerning the term “person.” Once the traditional meaning of that term is clarified,
I discuss the terms used to name that person, as confusion on those terms can
spell havoc for our conceptual clarity in these discussions.

2.1 The Definitions of the Terms “Hypostasis” and “Person”


Jesus Christ is referred to as both a person and a hypostasis in the conciliar texts.
To give just one example, as Cyril says in his third letter to Nestorius, accepted
at the Council of Ephesus,

Why should he who submitted himself to voluntary self-emptying for our


sake, reject expressions that are suitable for such self-emptying? All the
expressions, therefore, that occur in the gospels are to be referred to one
person, the one enfleshed hypostasis of the Word. For there is one Lord Jesus
Christ, according to the scriptures. (Tanner 1990, 56)

What, though, is meant by these terms, “hypostasis” and “person,” such that
Christ counts as exactly one hypostasis and exactly one person?
Elsewhere (Pawl 2016d, 32), I have followed the work of Marilyn McCord
Adams (2005, 37) and Alfred Freddoso (1986, 49) in analyzing the term “hypos-
tasis” (in the Latin translation, “supposit”), following the Medievals, to mean:

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
6 Philosophy of Religion

Supposit X is a supposit (hypostasis) if and only if x is


(Hypostasis) a complete being, incommunicable by identity,
not apt to inhere in anything, and not sustained by
anything.

The first clause of the right side of the biconditional is intended to preclude parts of
supposits from being supposits themselves – my hand is not itself a supposit, to use
Aquinas’s example (ST III q.16 a.12 ad.2).5 Wholes, not parts of wholes, are
supposits. The second clause is included for technical Trinitarian reasons. In
brief, the question arose whether the divine nature itself is a hypostasis, in addition
to the three divine persons. The answer had to be “no” for traditional Christianity,
since there are only three, not four, persons. This second clause is meant to
safeguard that “no” answer. To be communicable by identity is for the very same
thing to be both had by one thing and given to another without the former ceasing to
have it. Things that are communicable in such a manner aren’t themselves suppo-
sits. The third clause is meant to preclude anything accidental from counting as
a hypostasis. This was a worry, since some views of the Eucharist claim that
accidental forms could exist without inhering in substances.6 Such non-inhering
accidents are not supposits in their own right. The fourth is most relevant to our
purposes. The Medievals intended the notion of being “sustained” here to preclude
the human nature of Christ, which is sustained by the person, from itself counting as
a hypostasis.7 That human nature exists in the Word, depending on the Word for its
existence. Things that depend upon others in the way that the human nature
depends upon the person are not supposits in their own right.8 One can find similar
understandings of hypostasis throughout the Christian tradition.9
With this notion of hypostasis in hand, forming a notion of personhood is an
easy matter. A person, on the traditional understanding of the term, is
a hypostasis that has a rational nature.10 All supposits have some nature or
other. Dogs have their own natures, slugs have their own, humans have their
5
This citation ought to be read: Aquinas’s Summa Theologica, third part, question 16, article 12,
response to objection 2.
6
For discussion of this point, see Pawl (2012).
7
See also de Aldama and Solano (2014, 42–53) on this point.
8
It is true that, in another sense, everything created is sustained in existence by God on the
traditional Christian view. That is a different sense of the term “sustained.” The Medievals were
not saying that nothing created is a supposit.
9
See, for instance, Carlson (2012, 129, 259), Geddes (1911), Gorman (2000b; 2017, chap. 1),
John of Damascus (1958, 20, 56), Pohle (1911, 222), Rebenich (2002, 73), Salano and de
Aldama (2014, 42–43), and Tanner (2001, 32).
10
To see other discussions of this traditional conception of “person”, see Adams (2005, 23–24),
Aquinas (2012, 13), Carlson (2012, 204), Cupitt (1977, 135), Ferrier (1962, 81), Flint (2012,
189), Geddes (1911), Gorman (2011, 430), Lonergan (2016, 387–389), Pawl (2016d, 32–33;
2019b, 22), Pohle (1911, 224), Sturch (1991, 269–274), Turcescu (2005), Twombly (2015,
57–60), Wesche (1997, 95, 126), and C.J.F. Williams (1968, 517).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 7

own, etc. But only some of those natures are rational. And so only some
supposits count as persons.
Interestingly and importantly, the notion of person used in these discus-
sions is not a modern notion of personhood, called by some in the debate
a Lockean notion of personhood (Cf. Pohle 1911, 226). Such a notion has it
that personhood is, as Carlson (2012, 204) defines it, “An individual who
manifests the developed traits and abilities associated with human, personal
life (e.g. self-awareness, deliberate choice and action).” The reason this
modern definition will be insufficient for the Christological (and
Trinitarian) contexts becomes more explicit in the discussion of the humanity
of Christ in Section 4. In brief, though, the traditional view is that Christ had
a human element – a body/soul composite – that some argue fulfilled the
conditions for being a person in this modern sense of the term. This may lead
some to believe that there are two persons – “persons” in the sense relevant to
the doctrine – in the Incarnation, something the conciliar texts and traditional
Christian orthodoxy adamantly deny. This is just one example of how ignor-
ance of the historical meaning of the technical terms can cause confusion
concerning the traditional doctrine. In Section 2.3, we see another instance of
confusion over terms.

2.2 One Person of the Trinity


Jesus Christ is a person – a hypostasis of a rational nature. Not only that, he is
one person of the Blessed Trinity on the orthodox Christian view. Third
Constantinople says as much, claiming that the council, along with the previous
five councils, “professes our lord Jesus Christ our true God, one of the holy
Trinity” (Tanner 1990, 127). This person did not come into existence within the
Virgin Mary (a view Cyril calls “absurd and stupid” in the conciliar texts
[Tanner 1990, 42]). Rather, this person “existed before every age and is coeter-
nal with the Father” (Tanner 1990, 42). Since he is a member of the Trinity,
coeternal with the Father, he is a divine person.
Is Jesus a human person, as well as a divine person? This question has
typically been answered in the negative throughout Christian history, for com-
plicated reasons. Such an answer has caused confusion in some thinkers. For it
seems that one rightly counts as a human person if one is both human and
a person. Christ fulfills both criteria. Christ “became human” according to the
Nicene Creed. Moreover, as we’ve seen, he is a person. Why, then, this denial of
Christ’s being a human person? We can see the hesitancy here by looking at the
history of the term “human person” in the Christological discourse, combined
with a traditional prudential maxim from those debates.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
8 Philosophy of Religion

Concerning the history, the term “human person” was used to mean merely
human person. As such, speaking in this way was viewed as problematic by
orthodox proponents of the Incarnation. Philosophers will rightly note that
“human person” doesn’t logically entail “merely human person” by itself. But
philosophers will do well to remember that, in matters of preaching and
teaching, conversational implicature is at least as important as logical
implicature.
Concerning the traditional prudential maxim from Christological debates, the
idea was not to speak as the opponents spoke, for fear of leading astray the laity.
Aquinas, for instance, notes that while it is true that Christ is impassible in his
divine nature, one ought not to say so unqualifiedly, as it may appear to be
a word in favor of the view of the Manicheans, who thought that Christ had no
real body (ST III q.16 a.8 ad.2). The same prudential maxim would lead the
orthodox proponents of the Incarnation not to speak in ways that appear to
support the view of the Arians, who thought Christ was not a divine person in
the same sense the Father is. To make explicit that they were affirming neither
a Manichean nor an Arian view of Christ, some thinkers in the tradition took to
calling Christ a “Theandric” person. The English word “theandric” comes from
two Greek roots, which in the nominative are theos, for God, and anēr, for man
(which is andros in the genitive case).11

2.3 The Names of the Person


Many terms are used in the Christological writings: “Jesus,” “the Christ,” “the
Word,” “the Son,” “the Logos,” “the Second Person of the Trinity,” etc. There is
some disagreement about what exactly is referred to with each term. We can see
this most clearly if we focus on the name “Jesus Christ.”
Some people use “Jesus Christ” to refer to the assumed human nature alone
(cf. Holland 2012, 81; Marmodoro and Hill 2011, 13; and Sturch 1991, 122,
141).12 Others use “Jesus Christ” to refer to the person of the Son, the very
person that exists eternally with the Father and Holy Spirit, and who assumes
human flesh from Mary (cf. Alfeyev 2012, 265; Pawl 2016d, 46–47; Jedwab
2011, 169; Rea 2011, 150; and Wesche 1997, 12–13). Finally, some use that
name to refer to a third thing, not the assumed human nature alone, and not the
person of the Son or the Word. Rather, they use the name “Jesus Christ” to refer
to a compound – in some sense of compound – of that person and the human
nature into a “larger” entity.13 The view here is that “Jesus Christ” names
11
I thank Jonathan Rutledge for discussion of the Greek roots here.
12
In doing so, they are understanding the nature in the concrete sense discussed later in Section
4.1.2.
13
See Crisp (2011; 2016, chap. 6), Leftow (2011, 321), and Turner (2019, n. 5).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 9

a contingent entity, not itself a person, but that has a person as a part (see section
5.3 for more on this view, there called “Model A”).
In this Element, I use the terms “Jesus Christ,” “Word,” “Second Person,”
“the Son,” and all the rest as co-referring personal terms. They refer, in my
usage, to one and the same person. I have argued for this usage of the terms
elsewhere on conciliar grounds (Pawl 2016d, 46–47), grounds Flint (2011,
81n.17) shares. Against the first view, that “Jesus” names the human nature
alone, we do well to recall that the councils call Jesus “true God” and “one of the
Holy Trinity” (Tanner 1990, 127). No human nature, though, is one of the Holy
Trinity or truly God. Against both the first and third views, the councils call
Jesus a hypostasis and a person, as we saw in Section 2.2. The Word too, though,
is a person. If we look back to the truth conditions for being a hypostasis in
Section 2.1, we see that no hypostasis can have a distinct hypostasis as
a component, since each hypostasis is a complete being in its own right. And
so the Word and Jesus cannot be related to one another as component to whole,
since they are both referred to by “person” and “hypostasis.” There must be two
distinct persons, then, or the names must refer to the very same thing. If distinct
persons, we have a nonorthodox view sometimes labeled “Nestorianism.” If the
terms co-refer, then the naming convention I use in this Element is the correct
one.14
It is true that the name “Jesus Christ” is bound up with the Incarnation, such
that had the Word not become incarnate, he wouldn’t have had that name.
Presumably, “the Second Person” is not contingent upon the Incarnation in
the same sense. Even so, that doesn’t show that the two terms do not co-refer.

2.4 Conclusion
In summary, the Word, the Son, the Second Person of the Trinity, is a person –
a hypostasis of a rational nature. That very person became human and is the one
and only divine person to have become incarnate.15 As I use the term in this
Element, “Jesus Christ” is the name of a person, and that person is the very same
person as the Word.
14
See Section 5.3 for more discussion of the relation between Jesus and the Word.
15
It is an interesting question whether the other two divine persons could have become incarnate.
For more on that question, see Adams (1985; 2005; 2006, 198–199; 2009, 241;), de Aldama and
Solano (2014, 63), Aquinas (ST III q.3), Arendzen (1941, 161), Baker (2013, 47), Bonting
(2003), Brazier (2013), Craig (2006, 63), Crisp (2008; 2009, chap. 8), Cross (2005, 230–232),
Cuff (2015, 366–371), Davies (2003), Fisher and Fergusson (2006), Flint (2001, 312; 2012,
192–198), Freddoso (1983; 1986), George (2001), Gondreau (2018, 145–150), Gorman (2016),
Hebblethwaite (2001; 2008, 74), Jaeger (2017), Kereszty (2002, 382), Kevern (2002), Le
Poidevin (2009a, 183; 2011), Mascall (1965, 40–41), Morris (1987, 183), O’Collins (2002,
19–23), Pawl (2016a; 2016c; 2019b, chaps. 2 and 3), Pohle (1913, 136), Schmaus (1971,
241–242), Sturch (1991, 43, 194–200), and Ward (1998, 162).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
10 Philosophy of Religion

3 The Divinity of Christ


The previous section discussed the person of Christ. This section is the first of
two that take up the question of what Christ is: fully God and fully human. This
section focuses on his divinity, whereas the next section focuses on his
humanity.

3.1 The Divine Nature Itself


According to traditional Christian teaching, there is only one divine nature, also
referred to as the divine substance or the divine essence (Tanner 1990, 114).
That one divine nature is possessed (in some sense) by the three divine persons.
The claim that there is only one divine nature is intended to protect against
tritheism; the distinction between the three persons is intended to protect against
modalism (the view that the Father, Son, and Spirit are merely different ways
that the one divine person chooses to manifest himself to the world).16

3.2 The Attributes of the Divine Nature


Scholars have presented various views concerning the attributes of the divine
nature. This section canvasses the main views in the contemporary debate.

3.2.1 The Classical View

On one understanding, the divine nature has the attributes of classical theism.
On this theory, the nature is impassible and so unable to be causally affected.
The nature is immutable and so unable to change. It is atemporal and so outside
of time. And it is simple so without any ontological complexity.
Parts of this view receive support from the conciliar texts.17 For instance,
concerning immutability, the Council of Ephesus teaches that Christ “is
unchangeable and immutable by nature” (Tanner 1990, 51). The same council
teaches that “those are quite mad who suppose that ‘a shadow of change’ is
conceivable in connexion with the divine nature of the Word” (Tanner 1990,
72). Concerning impassibility, Chalcedon says that it is an error to claim that
“the divine nature of the Only-begotten is passible” (Tanner 1990, 84).
Moreover, Chalcedon “expels from the assembly of the priests those who dare
to say that the divinity of the Only-begotten is passible” (85–86).
Evidence for atemporality in the councils is less clear. Leo writes in his Tome
to Flavian, part of the accepted documents from Chalcedon, that “whilst

16
See Hasker (2019) for a recent discussion of the divine nature.
17
To see the conciliar evidence in much greater detail, as well as evidence from later Catholic,
Orthodox, and Protestant sources, see Pawl (2016d, chaps. 8, Section II).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 11

remaining pre-existent, [the Son of God] begins to exist in time [ante tempora
manens esse coepit ex tempore]” (Tanner 1990, 79). The Second Council of
Constantinople includes an anathema that declares accursed those who will not
say that the Son has a nativity from the Father achronos, that is, outside time.
Neither of these texts proves that the divine nature is atemporal. But they do
give some reason to think that the Son is atemporal, and he clearly is not
atemporal due to his human nature. So there is reason to think that it is due to
his divinity itself that he is atemporal. Concerning simplicity, there is no explicit
affirmation from the first seven councils that the divine nature is simple. Later
councils and confessional statements, though, Catholic, Orthodox, and
Protestant, include the divinity’s simplicity as a teaching.18

3.2.2 Nonclassical Views

There are various ways one might understand the divine nature such that it
doesn’t have the robust attributes classical theism ascribes to it. One standard
term for the most-attested contemporary school of thought concerning the
divine nature that is contrary to classical theism is “theistic personalism.”
I avoid that term here, primarily because the rivalry between these two
approaches to God spills out into other topics besides the divine nature, for
instance, how God creates and whether God is a person or merely personal.19
Those topics aren’t germane to the purposes of this Element and so would be an
unnecessary rabbit trail to follow.
Another theory that may appear relevant here is kenotic Christology, which is
the topic of Section 7.3.4. Briefly, kenotic Christology claims that Christ
emptied himself of some of his divine attributes when he became incarnate.
For instance, he went from being omniscient to being limited in knowledge, or
omnipresent to being located only at a certain place. Such a view has an oblique
relation to the theory of the divine nature, for such a view requires that
omniscience is not an essential feature of divinity. Were it essential, Christ
would cease being divine when he lacks knowledge during the Incarnation.
Thomas Morris draws out the implications for divinity explicitly when he notes
that to be divine, on the kenotic view, wouldn’t require omniscience, but rather
omniscience-unless-incarnate.20 Likewise for other attributes that cause

18
For examples, see the Catholic Fourth Lateran Council (Tanner 1990, 230–232), the first three
decrees of the Orthodox Third Hesychastic Council, and the Reformed Belgic Confession of
Faith, Article 1.
19
For a discussion of the distinction between classical theism and theistic personalism, and to see
where the latter term was coined, see B. Davies (2004, chap. 1). For more on whether God is
a person or personal, see Page (2019)
20
See Morris (1987, 99); see also Forrest (2009, 130–131).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
12 Philosophy of Religion

Incarnational worries – they all get an “unless” modification. Even still, one
might think that the divine nature itself has the attributes of classical theism,
even if the persons can freely choose to be empty of those attributes, and so
kenotic Christology does not imply that the divine nature itself is not simple,
atemporal, immutable, and impassible.21
The main alternatives to classical theism seldom reject the listed attributes
wholesale. For instance, many keep the language of divine eternity, but under-
stand eternity as implying everlastingness, and not atemporality.22 Or one might
deny that God is unchanging metaphysically speaking, but retain that God is
unchanging morally speaking.23 Likewise, God might be causally affectable,
but unable to be caused to suffer psychologically.
These are common alternatives in the philosophical discussion of the
Incarnation. They do not, though, sit easily with the conciliar texts. If Christ
was already existing in time everlastingly, in what sense could he go from being
preexistent to beginning to exist in time, as we saw Chalcedon teach in Section
3.2.1? Moreover, we can tell the sense the councils give to the attributes by
seeing the work they intend the attributes to do. They use immutability, for
instance, to preclude ontological change in the divine nature, not merely to point
out moral constancy (Tanner 1990, 51).24

3.3 A Comparison of the Conceptions of Divinity with


Respect to the Incarnation
Depending on one’s conception of the divine attributes, the philosophical
problems that arise from the Incarnation will be different. To see how, consider
the philosophical objection that is the focus of Section 7.
Section 7 focuses on the objection that nothing can be both God and human,
since to be God it would have to be a way that no human could be. If one denies
that God is as the classical conception requires, then one needn’t answer the
question of how Christ was both, say, impassible and passible during the
Incarnation. Generally speaking, the problem of Section 7 can be viewed as
a “metaphysical size-gap,” as Marilyn McCord Adams termed it (2000, 103;
2004). The more one shrinks that gap by denying classical attributes of God, the
fewer problems one will have with rendering the Incarnation coherent.

21
Such a view would hold that while the self-emptying of the person requires change in the person,
it doesn’t require change in the divine nature. I leave the spelling out of the mechanics of such
a view to its proponents.
22
See Deng (2018) and Mullins (2016). 23 See Dorner (1994) and Swinburne (1993, 219).
24
For a much fuller discussion of the use of immutability in the conciliar texts, see Pawl (2016d,
107–114).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 13

3.4 Conclusion
This section explicated various views of the attributes of the divine nature.
According to Conciliar Christology, that nature is immutable and impassible.
A case could be made for its atemporality from those texts too, but such is not
explicit, as are impassibility and immutability. The section also discussed how
the rival conceptions of divinity affect the problems that face a philosophy of the
Incarnation.

4 The Humanity of Christ


The account of Christ’s humanity is vexed in the contemporary philosophical
discussion. Is it something abstract, like a shareable property, say, the Platonic
form of Humanity? Or is it something concrete, particular, and individual, say,
a flesh-and-blood composite? If the latter, what are its component parts? Does it
have both a body and a soul? Just one of those two? And if it has a soul, is the
Word playing the role of that soul, such that there isn’t a new, distinct, created
soul at the first moment of the Incarnation? If the human nature was a concrete,
created body and soul composite, how could one support the orthodox claim
that it wasn’t a person? Did Christ assume a fallen human nature?
Scholars disagree concerning the answers to each of these questions. This
section presents the contours of the debate, different packages of answers to the
questions, and reasons for thinking that some of these answers are incompatible
with Conciliar Christology.

4.1 The Human Nature Itself


Traditional Christian teaching contends that Christ assumed (a) human nature.
But what’s (a) human nature?25 The two main views, which are the focus of this
section, are that it is either an abstract, shareable thing or a concrete, individual
substance of the same type as your or my nature. There are, of course, other
notions of what the nature is. Michael Rea (2011, 149) presents a theory on
which the human nature is a power. That power is shared by all humans and is
a feature of humans, so in that sense, the view Rea considers is similar to the
abstract view. On the other hand, that power is causally efficacious and located
in space and time, and so is similar to the concrete view in those respects.
In what follows, I focus on the two main views of the human nature of Christ:
the abstract view and the concrete view.26

25
For an excellent, prolonged discussion of this question in the medieval context, see Adams
(1999).
26
For discussions of these two views of nature, see Crisp (2007b, 41), Dalmau (2016, IIA:68),
Dubray (1911), and Plantinga (1999, 184). For a discussion of the historical reception of the term
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
14 Philosophy of Religion

4.1.1 The Abstract View


On the abstract view of human nature, the human nature is a property (or
a collection of properties) the instantiation of which is both necessary and
sufficient for being a human. Insofar as you and I both instantiate (have,
possess) that shareable feature, you and I are both human.
On the abstract view, in becoming incarnate, the Son entered into an instanti-
ation relationship with Humanity, and so fulfilled the necessary and sufficient
conditions for being a human. He was, then, a true human, in the same sense that
I’m a true human.

4.1.2 The Concrete View


On the concrete view of human nature, a human nature is a concrete, particular
thing. It is, as the ecumenical councils sometimes refer to it, “flesh enlivened
by a rational soul,” a “holy body rationally ensouled,” and “human flesh
which is possessed by a rational and intellectual soul” (Tanner 1990, 41, 44,
115). On this view, the nature is not a shareable thing. It is composed of
particular flesh and bones. You and I don’t have the very same nature in this
sense of the term. Instead, a nature is “a principle of operation . . . active
through its powers – it causes things to happen in the real world” (Baker 2013,
37). This is a typical perennial understanding of the term “nature”; the nature
is the origin (natus) of the activity of a thing. “Nature,” in this sense, refers to
the same thing as “substance” and “essence” (another word with a perennial
set of meanings more broad than the contemporary analytic understanding),
but picks it out as the origin of action, and not as a subsisting thing (as
“substance” does) or by its definition (as “essence” does; cf. St. Thomas
Aquinas, De ente et essentia, para. 10).
On the concrete view, in becoming incarnate, the Son united a concrete nature
to his divine nature (in a relation known as the hypostatic union). Having a real,
flesh-and-blood concrete nature, he was, then, a true human, in the same sense
that I’m a true human.

4.1.3 A Comparison of the Abstract and Concrete Views

Readers will note that the abstract and concrete views of human nature are as far
apart, ontologically speaking, as two views can be. An abstract nature like
Humanity is typically viewed as shareable by many individuals, not located in
space or time, unable to enter into causal relations, and possessed by substances

“nature,” and a listing of which thinkers thought what about natures, see Pawl (2016d, chap. 2
section II.b; 2019b, chap. 1 IV.b).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 15

as a feature. A concrete nature like Christ’s “holy body rationally ensouled” is


on the contrary not shareable in the way a universal like Humanity is, located in
space and time, able to enter into causal relations, and not possessed as
a feature – on the concrete nature view, the human nature is not related to the
Word as a property to a substance.
Even if they are ontologically miles apart, we still find each view assumed or
defended by major thinkers in the Christian tradition. Nevertheless, as I have
argued elsewhere, the texts of Conciliar Christology offer strong reason to
affirm the concrete account, for at least two reasons, which I have elsewhere
dubbed the Cyrillic and Leonine arguments, based on where one finds the
evidence for the arguments most explicitly asserted.27
The Cyrillic argument begins with the ways the nature is referred to in the
councils. As noted in Section 4.1.2, the nature is referred to as a “holy body
rationally ensouled.” No abstract object, though, is such a thing, so the nature is
best understood as a concrete object.
The Leonine argument begins with the predicates the conciliar texts say of
Christ’s human nature. The councils say of the human nature of Christ that it can
hang on a cross, be pierced, weep, feel pity, and say, “the Father is greater than I”
(Tanner 1990, 80–81). No abstract nature could do or undergo any of those
activities, so the nature is best understood as a concrete object.
Even if the two views of human nature are radically different, there’s a sense
in which there is little disagreement between the entailments of the two views.
To see this, one can coin a neutral term, say, “human element,” meaning the
ontological component parts of a human. For some thinkers, this human element
will be wholly material.28 For hylomorphic thinkers, it will contain both body
and soul. In any case, no matter the underlying ontology of the human person,
the traditional Christian will affirm that Christ had a human element. For the
abstract nature theorist, having such an element will be a necessary condition
for instantiating Humanity. For the concrete theorist, that human element will be
nothing else than the human nature of Christ. In a similar way, Brian Leftow
(2004, 279) has argued that the Word becomes incarnate in a concrete human
nature if and only if the Word begins exemplifying the abstract human nature.29
That claim is not required for the point here; rather, all that is needed is the
recognition that both views require Christ to take up a human element.

27
See Pawl (2016d, chap. 2 section II.b; 2019b, chap. 1 IV.b). In the following arguments I make an
assumption that either the concrete view or the abstract view is correct. These two views are
surely the two dominant views in the literature.
28
Both those who think the human element is wholly immaterial and those who think it is wholly
material will need to explain conciliar passages that refer to it as having both a body and soul.
29
See also Hasker (2016, 434–435), Leftow (2004, 278), and Marmodoro and Hill (2008, 101).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
16 Philosophy of Religion

Both abstract theorists and concrete theorists posit a human element for Christ.
Another – traditional – name for that human element is “human nature.” So, for
the remainder of this Element, I use the term “human nature” to refer to the
human element of Christ. If the need arises to refer to the abstract nature, in
order to avoid equivocation, I refer to it as “Humanity.”

4.2 The Parts of Christ’s Human Nature


Christ’s human element, the concrete human nature of Christ, is typically
thought to have parts. As quoted in Section 4.1.2, Conciliar Christology lists
some of those parts as a body and a rational and intellectual soul. It included
hands and feet that were pierced, blood, and all the other component parts of
a typical human being. At least that’s the orthodox view. There are other views,
though, concerning Christ’s human nature.

4.2.1 The Body of Christ


The view that Christ lacked a body is traditionally referred to as “Docetism”
(from the Greek word meaning “to appear”). On this view, Christ merely
appeared to have a human body or to be a human. Such a view makes little
sense of the scriptural passages concerning Christ’s physical activities.
Moreover, Christ sometimes seems to act in ways intended to show that he
really does have a body, as when he asks for some fish to eat after the resurrec-
tion (Luke 24:41–43), or when he allows Thomas to put his hand into his side
(John 20:27).
Another extreme view with respect to Christ’s body argues that the person of
the Word transformed from being immaterial to being a human material body, or
a human animal. Kevin Sharpe (2017, 118) writes that “the Son’s becoming
human was a matter of his being transformed into a human and not merely his
assumption of a concrete human nature.”30 Such a transformational view has
a dubious pedigree in Christology. As Cyril says in his second letter to
Nestorius, accepted at the Council of Ephesus,

For we do not say that the nature of the Word was changed and became flesh,
nor that he was turned into a whole man made of body and soul. Rather do we
claim that the Word in an unspeakable, inconceivable manner united to
himself hypostatically flesh enlivened by a rational soul, and so became
man and was called son of man. (Tanner 1990, 41, emphasis added)

30
For discussion of this or similar views, see Arcadi (2018, 4), Crisp (2009, chap. 7), Jaeger (2017),
Leftow (2015), Lim (2019), Merricks (2007), Turner (2017) and van Horn (2010).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 17

Similarly, Leo says in his Tome, “the Word is not the same thing as the flesh”
(Tanner 1990, 81). Both views, then – that Christ lacked a body and that he was
transformed into a human being – are precluded by Conciliar Christology.

4.2.2 The Soul of Christ


The view that Christ lacked a created, human soul is traditionally referred to as
“Apollinarianism,” after its most famous advocate, Apollinaris (the younger) of
Laodicea. Some thinkers, seeing that if Christ had a created, human soul, then
he’d have two centers of cognition (one divine, one human), worried that there
would be two distinct persons there, contrary to biblical witness and traditional
teaching. In order to safeguard Christ’s having a soul, but also to remove the
alleged problems of his having two faculties by which to reason or will, these
thinkers had the person of the Word himself play the role of Christ’s soul. So, on
such a view, Christ has a soul (the Word), yet only one set of intellective powers.
One finds such a view in Moreland and Craig (2003, 611), where they note
that the councils preclude it, but claim that the Scriptures allow it.31 Craig has
called this view “Neo-Apollinarianism,” and said that it includes the claim that

The soul of the human nature of Christ is the second person of the Trinity, the
Logos. The human nature of Christ is composed of the Logos and a human
body. (Harris and Craig n.d.)

What ought one to make of these views, where the divine person is the human
soul of Christ?
The main objections to such a view are the following. First, the soul is that
which makes a human to be human. Lacking a human soul, then, Christ
wouldn’t be truly human. Second, consider the maxim of St. Gregory of
Nazianzus that what is not assumed is not healed. Were there no assumed
human soul, then, there would be no healing of our human souls by the
Incarnation. For these and other similar reasons, Apollinarianism was declared
a heresy at the second ecumenical council, the First Council of Constantinople
in AD 381, and reaffirmed as a heresy many times over in later councils.
Concerning whether Neo-Apollinarianism is distinct from Apollinarianism
in anything more than name, note how a church council in Rome in AD 382
condemned Apollinaris’s view:

We pronounce anathema against them who say that the Word of God is in the
human flesh in lieu and place of the human rational and intellective soul. For,
the Word of God is the Son Himself. Neither did He come in the flesh to

31
Rea (2011, 149–150) presents without endorsement such a view.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
18 Philosophy of Religion

replace, but rather to assume and preserve from sin and save the rational and
intellective soul of man. (as quoted in Sollier 1907)

Neo-Apollinarianism is the doctrine that, eight years before his death,


Apollinaris was condemned for teaching.
Recall the quotation at the end of Section 4.2.1. It claims that the thing united
to the Son in the Incarnation was “flesh enlivened by a rational soul.” If the soul
were the Logos, the person of the Son, then the Son unites to himself flesh
enlivened by himself. Such a doubled union – Son to flesh, then Son to (Son and
flesh) – is unheard of in Christological discussions.32
Finally, recall the worry that partially motivates acceptance of
Apollinarianism – if there were two centers of intellection, then there’d be
two persons there.33 Such a worry rightly attempts to safeguard the unity of the
person of Christ, which is a necessary condition of traditional Christology. At
the same time, though, another necessary condition of traditional Christology is
dyothelitism, the claim that Christ had two distinct wills, a human will and
a divine will. For instance, the sixth ecumenical council, the Third Council of
Constantinople, includes an exposition of faith that affirms:

[T]he difference of the natures [is] made known in the same one subsistence
in that each nature wills and performs the things that are proper to it in
a communication with the other; then in accord with this reasoning we hold
that two natural wills and principles of action meet in correspondence for the
salvation of the human race. (Tanner 1990, 129–130)

Here the traditional texts teach that there are two natures, and they are known
because each has its own, distinct will. As such, even if one worries that two
wills implies two persons, the response of denying two wills is not an open
option for the proponent of Conciliar Christology.
Some have attempted to read this passage as consistent with Christ having
a single will that counts as both divine and human. In this attenuated sense,
though numerically one, Christ has two wills. Such a reading would allow the
proponent of traditional Christology to avoid the two-person worry and yet
maintain consistency with the councils. Crisp (2007b, 59–60) and I (Pawl
2016d, 19–20) argue that this reading is inconsistent with other claims the
councils make concerning Christ’s wills. The divine will leads the human
will, which is subjugated to it; each nature has a will and operates through it,
etc. Such claims are difficult to parse if there is only one will.

32
For more on the topic of a double union, see Pawl (2016d, 48–50).
33
For more on the topic of whether two minds or wills implies two persons, see Pawl (2016d, chap. 9).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 19

Given that Conciliar Christology does teach the unity of the person, along
with the duality of natures, each with its own will and operation, how does the
Christian avoid the objection that there are, in fact, two persons in Christ?

4.3 Was Christ’s Human Nature a Person in Its Own Right?


The answer to the question in this subheading is “no” for all traditional Christian
thinkers, as the opposing view, which often goes under the name
“Nestorianism,” was condemned at the Council of Ephesus in AD 431. The
human nature is not a hypostasis at all, and so not a person, since all persons are
hypostases. That said, the question still arose concerning the relation the human
nature bore to personhood. Was the human nature entirely non-hypostatic
(anhypostatic), or was it instead hypostasized in and by the person of the
Word (enhypostatic)? The orthodox answer was that the nature is personal
(enhypostatic), but it is itself neither a person (hypostatic) nor wholly imper-
sonal (anhypostatic).34 The reason given for these views, though, is different for
different schools of thought.
For those who believe the human nature to be an abstract entity, Humanity,
the answer is easy: no abstract, shareable thing is a person in its own right. In
fact, on the abstract view, it is hard to see how this perennial Christological
worry concerning the human nature of Christ could be taken seriously. But even
still, people who affirm an abstract nature believe that Jesus has a full compo-
nent of human parts, such as his body, hair, skin, blood, etc. Take all those
human parts together, which I referred to in Section 4.1.3 as Christ’s human
element: why isn’t that thing a person?
If, contrary to orthodox Christianity, one denies that Christ had a created human
soul, then there is again an easy answer. The full human element of Jesus was
without any second thing that operated rationally. Thus, there’s little motivation to
worry that the nature might fulfill the conditions for itself being a person. The word
“full” in “full human element” becomes questionable at this point, though.
Consider, then, those who think that the full human element, the concrete
human nature of Christ, had all the same types of ontological components that
yours or mine does. Suppose, as the councils say, it had its own intellectual
faculties, its own will and operation, as we saw in Section 4.2.2. In such a case,
why isn’t it a person?
By the medieval era, two main schools had emerged in answer to this
question.35 More than 100 years ago, Charles Dubray (1911) put the division
well:

34
For more on the enhypostatic/anhypostatic distinction, see Crisp (2007b, chap. 3).
35
For more on these two schools, see Hipp (2001, 471–518) and Pawl (2016d, 68–70).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
20 Philosophy of Religion

The Human Nature in Christ is complete and perfect as nature, yet it lacks that
which would make it a person, whether this be something negative, as
Scotists hold, namely the mere fact that a nature is not assumed by a higher
person, or, as Thomists assert, some positive reality distinct from nature and
making it incommunicable.

The next two subsections discuss these views.

4.3.1 The Thomistic View

One view, often associated with St. Thomas Aquinas, is that being a person
requires a certain component part. Christ’s human nature, on this view, lacks
that ontological component, and so fails to be a person.36
What exactly the part is differs in different views. Some call it a mode, others
an individual act of existence, others still something else. For the purposes of
responding to the charge that the human nature of Christ is a person in its own
right, the ontological category that the thing falls under is not as important as the
method of response. Christ’s human nature lacks an ontological component that
it would need to have in order to be a person, and so it fails to be a person.

4.3.2 The Scotistic View

The Scotistic view posits a negative condition for personhood. For a nature to be
a person, it has to be unassumed. Since Christ’s human nature fails that condi-
tion, since it is assumed, it does not count as a person. This view does not require
the supposition that your or my nature has a certain ontological doodad in virtue
of which we are persons. Instead, it merely requires our natures to be unassumed
for us to count as persons.

4.3.3 A Similarity between the Views


One might pause here to reflect on a similarity between these views. Given
orthodox Christology, as we see in the next section, there is something referred
to as a hypostatic union. St. Thomas Aquinas and many others view that union
as itself a created thing, something had, in some sense, by Christ’s human
nature. If that’s the case – if Christ’s human nature has a special ontological
component in virtue of which it is assumed, then we get similarity between the
Thomistic and Scotistic views. Letting “CHN” abbreviate “Christ’s human
nature,” consider the claim “CHN is a person” and its negation, “it is false
that CHN is a person.” The Thomist requires CHN to have an additional

36
Textual evidence indicatess, though, that Aquinas held a view much closer to the Scotistic view
discussed later in this Element. See Pawl and Spencer (2016, 148–149).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 21

ontological component for the first claim to be true. In the language of


truth makers, it is CHN plus the component that makes “CHN is a person”
true. Such a component is a necessary condition, and so in worlds where CHN
lacks a component, for instance, the actual world, “CHN is a person” is false.
The Scotist requires CHN to have an additional ontological component for
the second claim to be true. In the language of truth makers, it is CHN plus
the hypostatic union that makes “it is false that CHN is a person” true. Such
a hypostatic union is a necessary condition for the truth of that claim, and so in
worlds where CHN exists but lacks a hypostatic union, “it is false that CHN is
a person” is false – in other words, in such worlds, “CHN is a person” is true.
Both theories, then, require some ontological doodad in virtue of which one or
other of the contradictory predications about personhood is true.

4.4 Was Christ’s Human Nature Fallen or Unfallen?


This question is vexed.37 It is vexed in part because the word “fallen” is
understood in different ways in different traditions. Fallen from what, to what,
in what respect? In a diminished sense of the term, every traditional Christian
agrees that Christ’s human nature was weakened, insofar as, while it was still of
the very same type as Adam and Eve’s human natures, it lacked certain good-
making features of those natures in their original state. Many Christian thinkers
have claimed that Adam and Eve had certain preternatural gifts in the garden,
such as full control over their bodily passions and immortality (see Aquinas’s
Summa theologiae I, qq. 94–97 for more on these gifts). Jesus did not have all of
these gifts – for instance, he wasn’t immortal according to his humanity – and
so, in that respect, his human nature was weakened.
On the other hand, for some thinkers, to say that Christ’s human nature was
fallen is to say that, in virtue of it, Christ was potentially sinful. Oliver Crisp
(2007b, chap. 4; 2004) understands fallenness in this more robust way when he
claims that Christ’s having a fallen human nature would be catastrophic for
traditional Christian thinking, as fallenness implies having Original Sin –
a hereditary corruption or stain that results from the first sin of our ancestors
(see Crisp 2007b, 107).
The traditional answer to the question of whether Christ had a fallen nature
can be put in a way that garners broad consensus. Christ had a human nature that
was corruptible. That said, he did not have Original Sin, neither did he have the
actual ability to sin (for more on the ability to sin, see Section 6.1.2). There is
disagreement on whether such a state can rightly be called “fallen.” Some think

37
For more recent work on this question, see (2019), McFarland (2008), King (2015), McCall
(2019, chap. 4), and McFarland(2008)
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
22 Philosophy of Religion

fallenness requires Original Sin, others that it merely requires not having all the
preternatural gifts that Adam had. How much of a fall is required to count as
a fallen human nature is a question we need not answer at this point.
We find others drawing these distinctions in the understanding of fallenness
as well. Immediately before offering pages of quotations verifying the claim,
Alfeyev writes:

Many pronouncements of the Eastern fathers, including Irenaeus of Lyons,


Athanasius of Alexandria, Gregory of Nyssa, Cyril of Alexandria, and
Gregory Palamas, leave no doubt that these fathers considered Christ’s
human nature, with the exception of sin, to be like the nature of fallen man.
It is from the fallen, and not the first-formed Adam, that Christ inherited
a corrupted, mortal, impassioned nature. (Alfeyev 2012, 278)

He goes on in the same section to claim that, for the Orthodox, Christ did not
have personal sin, sinful inclination, or Original Sin.38

4.5 The Weaknesses and Infirmities of Christ’s Human Nature


On the traditional view, Christ had weaknesses and infirmities during life. As
Scripture notes, he was at times hungry, sad, anxious, thirsty, etc. St. John of
Damascus claims that such “innocent” passions were not lacking in him. Just as
traditional is the affirmation that Christ lacked sinfully disordered passions. For
instance, consider a sinful inclination to some deplorable type of action (the
reader can supply her own). Such deplorable inclinations evince a faulty moral
character, though not one that is necessarily blameworthy, depending on the
source of the inclination. Such dispositions were lacking in Christ, it is argued,
because to have them is to have a moral fault, and no one who is God could have
a moral fault. The relation between sinfulness, passions, and temptation is the
topic of Section 6.1.2.

4.6 Conclusion
This section explored various divisions in the conceptualization of Christ’s
human nature. One might think of it as abstract or concrete. If concrete, then
there are further choices to be made about the number of parts it has, and the
person’s relation to that nature. If the nature is concrete, with both a rational soul
and a body (as Conciliar Christology teaches), then the question of whether it is
a person must be addressed. Two reasons for denying the nature’s personhood
that this section canvasses are that it lacks some component part necessary for

38
Alfeyev (2012, 285) notes that the notion of “Original Sin” in Western contexts isn’t the same as
in Eastern contexts, and in neither sense did Christ have Original Sin.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 23

personhood (the so-called Thomistic view) or that it satisfies some condition,


assumption, incompatible with personhood (the so-called Scotistic view).
Concerning fallenness, one must specify what exactly is meant by that term.
If fallenness merely means that it doesn’t have some good-making qualities that
Adam’s pre-fallen nature had, then Christ’s human nature is fallen, since it was
mortal. If fallenness goes further and requires Original Sin, concupiscence, and
liability to actually sinning, then traditionally the human nature assumed by
Christ is not fallen in that particular sense of the term.

5 The Hypostatic Union


This section focuses on the union that holds between the two natures in the one
person of the Word. What can we say of the union? What features does it have?
What role does it play in the Incarnation?39

5.1 The Features of the Hypostatic Union


The conciliar documents contain more about what the union isn’t than about
what it is. It was not merely by “God’s will alone or good pleasure,” nor by “a
conjunction of dignity or authority or power,” nor by a mixing of natures, nor by
affection, nor a union of adoration or honor, nor a union as one finds “between
a man and his wife” (Tanner 1990, 41, 59, 115–116, 119). In uniting, the two
natures undergo “no confusion, no change, no division, no separation” (Tanner
1990, 86). Moreover, the union is, according to the councils, ineffable (Tanner
1990, 72, 117). It was brought about in “an unspeakable, inconceivable manner”
(Tanner 1990, 41). These claims about ineffability and inconceivability are not
meant by the conciliar fathers in such a robust sense that we cannot say or think
anything at all about the union. After all, they do say many things about it, as we
see in the next paragraph. Rather, they mean that it is impossible for us to
completely grasp or understand the union.40
So much by way of negation. What can be said of the union positively? The
hypostatic union is the union that holds, in the person of the Word, between the
divine and human natures, in virtue of which Christ is one person of two natures.
The relata of the hypostatic union are, on the divine side, the divine nature, and,
on the human side, the human nature, or its parts.41 Elsewhere, the councils refer
to the hypostatic union as a union of “subsistence” or “synthesis” (Tanner 1990,

39
For a more detailed discussion of the hypostatic union in the conciliar texts, see Pawl (2016d,
20–23).
40
For more on the meaning of ineffability, see Pawl (2020, sec. 2). For more on mystery and
philosophical reflection on the incarnation, see Pawl (2016d, 88–91; 2019b, 5–6).
41
For more on whether the human relatum is properly characterized as the full human nature or the
parts of the human nature, see Pawl (2019b, 109–113).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
24 Philosophy of Religion

115). The two natures are united, in the traditional language, in the person of the
Word, but the person of the Word is not, technically speaking, one of the relata
of the relationship. (The term used for the relation of the Word to the human
nature is assumption.) It is important that the union does not take a person on the
human side of the relation, as discussed in Section 4.3. Were it to do so, there
would be two different persons in the Incarnation, rather than one person who is
both God and human, as the Nicene Creed claims.
It is this fact – the fact that there is one person who is both divine and human,
having both divine and human natures – that grounds the Communication of
Idioms.

5.2 The Communication of Idioms


The Communication of Idioms (Communicatio Idiomatum) is a thesis about
language. It is the thesis that what we can say truthfully of the one person of
Christ in virtue of one nature, we can say when referring to Christ with a subject
term drawn from the other nature.42 For instance, it is true that Christ was born
of Mary. And that predicate, “born of Mary,” is true of him, not in virtue of
something that happened to the divine nature, but rather in virtue of something
that happened to his human nature, that flesh-and-blood reality. Christ, though,
is God, and is God in virtue of his divine nature, not his human nature.
Nevertheless, we can put together a true claim using terms apt of him due to
different natures, as when we say that God was born of Mary.43 This, of course,
is the very case that got Nestorius in trouble, since he wanted to affirm that while
Mary bore Christ, she did not bear God. Such a claim – that Mary bore Christ,
but she did not bear God – Cyril and others reasonably argued, would imply that
the person, Christ, is not a divine person. Additional examples of the
Communication of Idioms, both found in Aquinas, are “A man created the
stars” and “The God of glory was crucified” (ST. III q.16 a.4). In both cases,
a predicate term apt of the one person in virtue of one nature is paired with
a subject term apt of the person in virtue of the other nature to make a true claim
about the one God-man, Jesus Christ.
We find support for the Communication of Idioms in the conciliar texts. We
already saw one supporting text at the beginning of Section 2.1. We see another
at the end of Section 7.4.2. For a third, Cyril includes in his third letter to
Nestorius the following anathema:

42
For more on the Communication of Idioms, see de Aldama and Solano (2014, 170–171), Cross
(2019), Pawl (2016d, 23–27, 54–55, 62–65; 2019b, 17–19), and Pohle (1913, 186).
43
I am using the term “apt” to mean correctly assertible of. Some thinkers say a term is true of
a subject. That language is acceptable, too, but I prefer to avoid it, as I take truth to be a feature of
a complete proposition, not a feature of a term in a proposition.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 25

If anyone distributes between the two persons or hypostases the expressions


used either in the gospels or in the apostolic writings, whether they are used
by the holy writers of Christ or by him about himself, and ascribes some to
him as to a man, thought of separately from the Word from God, and others, as
befitting God, to him as to the Word from God the Father, let him be
anathema. (Tanner 1990, 59)

Here Cyril, in a text accepted at the Council of Ephesus, asserts that the
expressions ascribed to Christ as human are ascribed to the very same subject
as the expressions ascribed to him as God.
It should be clear that the Communication of Idioms doctrine as spelled out
here does not imply that a predicate truthfully said of one nature is thus
truthfully assertible of the other nature, or that a predicate truthfully said of
the person is thus truthfully assertible of either nature.44 From Christ’s being
crucified, we cannot derive that his divine nature was crucified. From Christ’s
human nature being pierced on the wood of the cross, as Chalcedon teaches (see
Section 4.1.3), we cannot derive that the divine nature was pierced.
Likewise, the Communication of Idioms does not undergird either of the
following two theses:

1. Any predicate truly said of either nature of Christ is also truly said of the
person of Christ.
2. Any predicate truly said of the person of Christ is truly said of at least one
nature of Christ.

Neither of these is true, we can see, since they have counterexamples. Christ’s
human nature is truly said to be “identical to a human nature,” and there is a time
before which it did not exist, but neither of these predicates is truly said of the
person of Christ. The person of Christ is not “identical to the human nature,” and,
though there was a time before he was human, there was no time before which
that person did not exist (the original Nicene Creed precludes such a claim with its
final anathema). Likewise, the second claim is subject to counterexample. One
predicate truly said of the person of Christ is that he is “one person in two
natures.” But neither the divine nor the human nature is “one person in two
natures.” We find, then, that neither of these theses survives scrutiny.
In fact, the notion of “communication” may be misleading here, even if it is
the traditional language. It is not that the term goes from the nature to the person,
as a migratory bird might go from one location to another. Rather, it is that the
incarnate person fulfills the ontological conditions that are needful in typical
cases to satisfy the predicate, and that very same incarnate person can be
44
Though some Lutheran theologians have taught such communication from nature to nature; see
Pohle (1913, 194–195) for one discussion of this.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
26 Philosophy of Religion

referred to by terms that refer to him in virtue of his other nature. So we can use
a subject term from one nature, since the very person referred to by that term
satisfies the truth conditions for the predicate. It should not be thought of as
a predicate or a property migrating from one subject, whether linguistic or
ontological, to another. Rather, it should be thought of in terms of fulfilling
truth conditions or not.

5.3 Models of the Hypostatic Union


Even if the hypostatic union is ineffable, as we saw that Conciliar Christology
claims in Section 5.1, one could still ask whether a viable model exists for it, or
a helpful analogy.
Thomas Flint (2011, 71–79) draws a division between models of the incarna-
tion that has become standard.45 On Model T:

In becoming human, the Son or Word of God (whom I’ll label W) takes on
CHN as a part. This assumption results in a Son who combines both his
original divine substance (D) and his created human nature (CHN). (Flint
2011, 71)

On Model T, the Word comes to have an additional part (or part-like thing): the
human nature.
On the second model, Model A:

The Son unites himself to CHN in the Incarnation. But the composite thus
formed is not the Son. The Son remains simply one part of the composite entity
that results from his assuming a human nature. That composite entity, which
(following Scotus and Leftow) we can call Christ, is a contingent thing,
composed of another contingent entity (CHN) and of a necessary one (the
Son). (Flint 2011, 79)

On Model A, a contingent entity called “Christ” comes into existence, which is


itself composed of two things – the preexistent Word and the created human
nature. Is there reason to favor one view over the other?
There are good reasons to prefer Model T to Model A. First is a reason based
on the proper relata for the hypostatic union. What two things are united in the
union? The councils say that “a union of two natures took place,” and “one and
the same Christ, Son, Lord, only-begotten [is] acknowledged in two natures,”
and “two different natures come together to form a unity, and from both arose
one Christ, one Son” (Tanner 1990, 70, 86, 41, respectively). Model A has the

45
For discussion of these two compositional models of the Incarnation, see Crisp (2011; 2016,
chap. 6), Flint (2015), Hasker (2015), Leftow (2011, 321), and Turner (2019 n 5).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 27

person of the Son, not the divine nature, as one of the relata, while Model T has
the natures as the two relata. Score one for Model T.
Second is a reason from the nature of personhood. On Model A, a whole (the
contingent composite, Christ) has a person (the Word) as a part. But on the
traditional view of personhood, explicated in Section 2.1, a person cannot be
a part of a larger whole. Thus, Model A sits uneasily with the traditional notion
of personhood.46
We might also ask about helpful analogies in addition to models. The
Athanasian Creed (Denzinger 2002, para. 40) and the Council of Ephesus
(Tanner 1990, 52) both use the analogy that as a man’s soul indwells his body,
so likewise the Word indwells the human nature. Katherin Rogers (2010; 2013)
offers an analogy based on virtual realities. A child can play a video game and in
doing so enter and act in a virtual world through the character in the game;
similarly, the Son, through taking up a created human nature, can enter and act
in the created world.47

5.4 Conclusion
This section presented an account of what the hypostatic union is not, then some
discussion of what the union does. The main theological work the union does is
ontologically undergirding the union between the two natures in one person, and
so providing an explanation for the Communication of Idioms. The union has
been modeled in myriad ways, including models based on a mereological
understanding of the union, a union of body and soul, and multiple sci-fi
analogies.
Having discussed the ontology of the person, the natures, and their union,
now we move on to discuss the activities of that person.

6 The Activities of Christ


This section focuses on the activity of Christ. What philosophical questions
arise from this activity? The main contemporary discussions of Christ’s activ-
ities might be grouped under three headings: with respect to his acts of will, his
acts of intellect, and his death.

6.1 Volitional Activities


According to traditional teaching, as noted earlier (Section 4.2.2), Christ has
two wills, one divine and one human. This duality of wills gives rise to

46
For a reply to these objections to Model A, see Crisp (2011, 52–56).
47
Hasker (2017) gives a similar example in terms of the sci-fi movie Avatar.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
28 Philosophy of Religion

questions about what each will can do and their interrelations. Was each will
free? Or, perhaps put differently: was the one Christ free with respect to each
will? Moreover, given that Christ had a human will, could he sin by means of
that human will? If not, as was traditionally believed, what implications does
that inability to sin have on his being tempted? In this section, I consider each of
these questions.

6.1.1 Was Christ Free?

The consensus in Christian thinking has been that Christ has two volitional
powers, a divine one and a human one, and that he can will freely with each
will.48 Such an answer has led to two types of objections: objections from the
interrelation of the wills and objections from volitional activities and
personhood.
Considering objections from the interrelation of the wills, one might well
wonder here what the relations are between the two wills of Christ. The texts of
Conciliar Christology are not silent on this issue. For instance, the Third
Council of Constantinople (681) says:

We proclaim equally two natural volitions or wills in him and two natural
principles of action which undergo no division, no change, no partition, no
confusion, in accordance with the teaching of the holy fathers. And the two
natural wills not in opposition, as the impious heretics said, far from it, but his
human will following, and not resisting or struggling, rather in fact subject to
his divine and all powerful will. (Tanner 1990, 128)

Here the human will is said to follow, be subject to, not resist, and not struggle
against, the divine will. As Richard Swinburne (1994, 198–199) and Garrett
Deweese (2007, 133), among others, wonder: how can a will be free in such
a state?49
The main question at this point is how strong the subjugation is. My
children might follow my will in a certain thing, say, cleaning their rooms,
without resisting or struggling. (I imagine that this is possible.) In doing
so, they don’t render themselves unfree. What is needed for the objection
to succeed is some reason to think that Christ’s human will is subjugated

48
See Pawl (2019b, 119–123), Pawl and Timpe (2016), and Wessling (2013) for more discussion of
this issue.
49
See Pawl (2019b, 126–131) for a detailed discussion of this objection and a reply. For a critical
discussion of Swinburne and Deweese, see Pawl (2016d, 219–220). For other discussions of
Christ’s freedom, see Gaine (2015b, chap. 7), Hebblethwaite (2008, 68), Hick (2006, 56),
Kereszty (2002, 392–396), McFarland (2007), McKinley (2015), Rogers (2016), and Sturch
(1991, 29, 167).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 29

in a much stronger way, as, perhaps, a hypnotized person’s will is subju-


gated to the hypnotizer.50
It is true that the examples just offered – a parent and child, a hypnotizer and
the one hypnotized – are examples of two different persons and their own
wills, and not examples of one person with two distinct wills. This is under-
standable, as no other example of a two-willed person exists that we can
employ. If there is some reason to think that having two wills in a single
person adds difficulties with respect to this objection from subjugation, above
and beyond those found in the two-person examples employed, the reason is
still to be given.
What of the second variety of worries, the personhood worries? If there
are two distinct wills, how are there not two distinct persons? The reader
does well to remember the traditional definition of “person” employed in
these debates. A person, as expressed in Section 2.1, is a supposit with
a rational nature. Since, as argued in Section 4.3, the human nature does not
fulfill the condition for being a supposit, that nature does not satisfy the
conditions for being a person. So there are not two persons in the relevant
sense of “person.”51

6.1.2 Could Christ Sin?

The consensus answer to the question of whether Christ could sin with either of
his wills is “no.” One might wonder how the inability to sin is consistent with
the freedom of both of Christ’s wills. Moreover, one might also wonder how he
could be tempted, if willing something sinful was impossible for him.
Concerning whether Christ could sin, contemporary literature has featured
extended discussion. The traditional teaching is that Christ is not only sinless
but also impeccable.52 If he is impeccable, does that render him not omnipotent,
owing to the fact that there is something – that is, sinning – which Christ cannot
do?53 Typically, the answer is “no”: being able to sin is not its own power,

50
For an argument for strong subjugation from a different starting point, see Werther (2005).
51
A reader might accuse this response of defining away the problem, rather than facing it head on.
Such a reader does well to remember that the concept of “person” I employ here is the traditional
understanding of it in these contexts. If the objector understands personhood differently, and so
sees this as a devastating objection, it is the objector who is defining terms advantageously.
I consider this worry at length in chapter 9 of Pawl (2016). See also Murray and Rea (2008,
82–84).
52
On the historical affirmation of impeccability, see Bavinck (2006, 314), O’Collins (1995, 281),
Pohle (1913, 214), Schmaus (1971, 259), and Weis (2003).
53
On the question of whether God can sin, see, for instance, Adams (2006, 75–79), Brümmer
(1984), Carter (1985), Funkhouser (2006), Garcia (1987), Gellman (1977), Leftow (1989),
Morris (1983) and Stump (2005, 102–7)
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
30 Philosophy of Religion

a power had by you but lacked by God.54 To sin is to use a power poorly, but
sinning is not its own power, and so, the typical answer goes, God’s inability to
sin is not the lack of some power had by others. As T. J. Mawson (2018, 42) puts
the point, the ability to sin is a liability, not a power, and so not a lack of power if
it is missing. God, as a perfect being, has all powers and no liabilities.
Concerning whether Christ could be tempted, again, the discussion is vast.55
The traditional view is that Christ could be tempted, as Scripture (e.g., Hebrews
4:15) and the councils say, yet Christ could not sin. How can one make sense of
such a conjunction of claims?
One might attempt to explain the temptation of an impeccable person in
various ways. First, one might distinguish between internal and external temp-
tation. Christ could be externally tempted – the devil might wave a good in front
of his nose – yet neither feel the internal, sinful impulses toward that temptation
nor be able to succumb to them.56 In such a case, it would be true to say he is
tempted (in the external sense) and such a temptation would not require his
ability to sin.
A second response is known as the epistemic response. To be tempted, on
this view, does not require the actual ability to sin; it merely requires that one
believe that one has the actual ability to sin.57 You might be duped into
thinking that you have the launch button for the Russian nuclear program.
And you might feel sorely tempted to make use of that button, even though you
don’t, in fact, have the ability to launch the missiles. This response requires
that Christ not know of his divine status and mission, such that he mistakenly
thinks that he is a person able to sin. Such a view is precluded by traditional
views of Christ’s knowledge, as we see in Section 6.2.1. Nevertheless, this
remains an open option for those not wed to the traditional views of Christ’s
knowledge.
A third response is to conceptualize temptation such that it doesn’t require
either the ability to sin or even the belief that one can sin. Rather, following

54
For more on this point, see McKinley (2009, 258), Morris (1986, 167), and Pawl (2019b,
151–153) and the works in the previous footnote.
55
On the questions of temptation and sin, see Canham (2000, 95), Couenhoven (2012, 406–407),
Crisp (2007b; 2007c), Davidson (2008, 395), Erickson (1996, 562), Fisk (2007), Gaine (2015b,
168–172), Hart (1995), King (2015, 73–76), McKinley (2009; 2011), Morris (1987, chap. 7),
Murray and Rea (2008, 82–90), O’Collins (1995, 283–284), Pawl (2019b, chap. 6), Pelser
(2019), Sturch (1991, 19–20), Swinburne (1994, 204–207), Ware (2013, chap. 5), Wellum
(2016, 459–465), and Werther (1993; 2012).
56
Both Aquinas (2013, bks. 4, lecture 1, pg 101) and Crisp (2007c, 178) argue that Christ’s
temptations have an external source. See Murray and Rea (2008, 86) and Pawl (2019b,
150–151) for more discussion of this view.
57
For discussions of this view, see Bartel (1995, 154–55), Hart (1995, 41), McKinley (2009,
239–243), Morris (1987, 147–148), O’Collins (1995, 283–284), and Pawl (2019b, 139–143).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 31

some psychological discussion in the literature on temptation, we might think of


temptation as follows:

Temptation: An affectively charged cognitive event in which


an object or activity that is associated with pleas-
ure or relief of discomfort is in focal attention, yet
the object of that desire conflicts with the person’s
values and goals. (Pawl 2019b, 144)58

Here, the affectively charged “pull” of temptation need not take something
sinful as its object. It might be the pleasure of the taste of a chocolate cake that
is in focal attention, yet that good is seen to conflict with a distal goal – the
goal, say, of losing weight via dieting. On this view, Christ could have the pull
toward food in the desert, or away from the discomfort of crucifixion, and yet
judge such pleasure or lack of discomfort as contrary to his goal of fasting or
his plan of atonement for sinful humanity. On this psychological view, then,
neither the ability to sin nor the belief that one can sin is a necessary condition
for temptation.
The tradition makes a distinction between “natural and innocent” passions
and those that are not.59 Some passions, such as fear, hunger, thirst, aversion,
sadness, and so on are blameless and natural to humans. These are passions
Christ might have, and they are a blameless basis for temptation, when under-
stood as outlined in this section. Other passions, as discussed by Oliver Crisp
(2007a, 176), are evil in themselves. Such vicious passions (e.g., urges to
sexually assault others) show an imperfect character. The view under discussion
here has it that Christ never had any such passions or appetitive pulls, and so no
internal temptation toward them.

6.2 Intellectual Activities


Just as Christ’s having both a divine and a human will gives rise to questions
concerning the states and interactions of those wills, so likewise his having both
a divine and a human intellect raises similar questions. Christ had two faculties
of knowing: a divine one and a human one. What did Christ know with respect
to those intellects? When did he know it? Did such knowledge raise new
problems for his freedom?

58
See Hoffman et al. (2012, 1319), Kavanaugh, Andrade, and May (2005, 447), and Milyavskaya
et al (2015, 678) for relevant psychological discussion. See Pawl (2019b, 143–51) for a detailed
explication of the psychological view of temptation. Something similar to this view of tempta-
tion is at work in Aquinas’s thought as well; see Gondreau (2018, 359).
59
See Alfeyev (2012, 283) and John of Damascus (1958, 323–324).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
32 Philosophy of Religion

6.2.1 What Did Christ Know, and When?


Since Christ, on the traditional Christian picture, had two intellects, one divine
and one human, the discussion of his knowledge is typically divided into the
knowledge that he had in virtue of his divine intellect, and the knowledge that he
had in virtue of his assumed human intellect.60 Concerning the divine intellect,
the traditional view is that Christ was omniscient. Anything there is to be known,
Christ knew it via his divine intellect, just as the Father and Holy Spirit do.
Concerning his human intellect, the traditional view is that Christ had three
types of human knowledge, in addition to his divine knowledge. He had the
beatific vision, as the blessed in heaven do, he had infused knowledge, and he
had acquired knowledge. Beatific knowledge is knowledge had by union,
participation, or similarity with God.61 It is qualitatively distinct from our
mundane forms of knowing. Infused and acquired knowledge are alike in all
ways except cause. The difference is that infused knowledge is imparted
directly by a divine act, whereas acquired knowledge is acquired as human
knowledge is typically acquired. If you learn my paternal grandmother’s first
name via my telling you, that’s acquired knowledge.62 If you learn it by God
“zapping it in,” that’s infused knowledge.
Aquinas, for just one example, posits this fourfold distinction in types of
knowledge of Christ (see ST III. q. 9). Why include all four types? The divine
type is easy to explain for non-kenotic thinkers. Christ retained his divine digni-
ties when incarnate. One divine dignity is omniscience. So he remained omnisci-
ent as he eternally was – in virtue of his divine intellect. Concerning the beatific
knowledge, the question of whether Christ had it is more vexed, even among
traditional Catholic thinkers.63 Typical arguments in its favor often go as follows.
Christ is the author of salvation for humans, which consists in the beatific vision.
But Christ cannot give a perfection that Christ does not have. So he must have it.
Another argument comes from fittingness: it is unfitting for the source of the
beatific vision to lack it.64 Concerning infused knowledge, Aquinas argues from
Scripture, for instance, Colossians 2:3, which says that in Christ “are hid all the
treasures of wisdom and knowledge.”65 Finally, what of acquired knowledge?

60
For discussions of Christ’s knowledge, see Denzinger (2002, 2184–2185, 2289), Gaine (2015a,
2015b, chap. 6;), Green (2017), Loke (2013), Margerie (1980), Moloney (2000), Rosenberg
(2010), Scarpelli (2007), and Wellum (2016, 454–459).
61
For more on the mode of beatific knowledge, see Aquinas’s Compendium of Theology, 216, ST
III q.9 a.3 ad.3, and ST III q.10 a.4 resp.
62
Marge.
63
For discussion of this controversy, see Weinandy (2004; 2014) and White (2005; 2016, chap. 5).
64
These are both sketches of arguments that Aquinas gives. See ST III. q.9 a.2 resp.; Comp. Theol.
216.
65
See ST III. q.9 a.3 sed contra.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 33

Recall that, on the traditional doctrine, Christ has a created human soul, the same
in type as yours or mine. Now, your soul and mine has natural activities, some of
which are intellectual. Were Christ to lack all acquired knowledge, a standard
activity and perfection of a human soul would be lacking in him. It would be
there, but “offline,” so to speak. This appears unfitting to many thinkers. So, like
typical humans, Christ could use his intellect to know things.
How can we explain the biblical passages that seem to require ignorance of
Christ, then? Consider, for instance, these two passages:66

Luke 2:52: And Jesus increased in wisdom and in stature, and in favor with
God and man.

Mark 13:32: [Jesus said,] “But of that day or that hour no one knows, not even
the angels in heaven, nor the Son, but only the Father.”

Historically, there are three main ways to deal with such passages.
First, a recent way of responding has been to suppose that Christ was really
ignorant of things like when he would return. Kenotic thinkers can claim that
Christ in no way knew when he would return, owing to the fact that he emptied
himself of his divine knowledge.
Second, some can claim that when Christ claimed lack of knowledge, he
meant that he lacked knowledge in his human intellect, not that he lacked
knowledge full stop.
Third, and more traditional, is the attempt to deny that such passages really
require ignorance. We find this variety of response in Cyril of Alexandria,
Gregory the Great, John Chrysostom, Augustine, Maximus the Confessor, and
Thomas Aquinas, to name just some thinkers who support this method of
response.67 Concerning the passage from Luke, Christ grew in appearance of
or renown for wisdom and favor with God, and also in acquired knowledge. This
growth in acquired knowledge is not a change from ignorance to knowledge;
rather, it is coming in a new way to know something already known. Concerning
the passage from Mark, these authors distinguish between the state of knowing
and the revelatory act of making known. Christ here, they claim, intends to say
that he does not make known the hour. As a parallel, some, like Aquinas (Comp.
Theo. 242), note that God says to Abraham after the potential sacrifice of Isaac,

66
See Archer (2017), Gaine (2015b, chap. 6), and Moloney (2000, 28–32) for additional discussion
of these passages.
67
For Cyril, see Bartel (1991, 35). For Augustine, see De Trinitate I.12.23. For Aquinas’s reference
to Chrysostom, see ST III q.10 a.2 ad 1. For Gregory, see Denzinger (2002, paragraph 248). For
Maximus, see Moloney (2000, 46). For Aquinas’s response, see his Compendium of Theology,
242. For Aquinas’s discussions of Luke 2:52, see: QDV q.20 a.1 sed contra; Comp. Theol. 216;
ST III q.7 a.12 obj 3 and ad 3; ST III q.12 a.2 sed contra, resp, and ad 3; his Commentary on John,
1:14, no. 264. For a critical discussion of Aquinas’s approach, see Maritain (1969, 50–54).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
34 Philosophy of Religion

“Now I know that you fear God” (Gen. 22:12). Aquinas argues that this does not
mean that God went from ignorance to knowing that Abraham feared him.
Rather, God was declaring it made known that Abraham feared him.

6.2.2 Christ’s Knowledge and Freedom


Supposing that Christ had knowledge by means of his human intellect of all future
contingent states – a view Aquinas claimed was held by everyone (Aquinas 1954,
QDV q.8 a.4 resp) – difficulties arise concerning freedom, both his and ours.
Concerning our freedom, some of the most common responses to the problem
of foreknowledge and freedom fail to generalize if Christ had such knowledge.
For instance, open theism, the view that the future is open and so not knowable,
fails as a solution. It fails because, on supposition, Christ does know the future. In
addition, the Eternity Solution, the view that God knows things atemporally and
so does not know them as foreknowledge, fails as well. For even if the divine
knowledge is atemporal, Christ’s infused human knowledge is in the past.68
A third potential response to this problem is to understand our freedom as
consistent with determinism, as the Compatibilists do. On such a view, deriving
that we couldn’t do otherwise than what we do at any time wouldn’t imply our
lack of freedom. A fourth response is to note that Christ’s infused knowledge is of
the same type as our knowledge. But what we know is made to be true by how the
world is, not vice versa. So likewise, Christ’s human knowledge is explained by
how the world is, not vice versa. You have control over Christ’s past human
knowledge, insofar as you have control over what you do, and it is because of
what you do that Christ knows what he knows, not vice versa.69
Concerning Christ’s freedom, there are three problems to consider. First is the
issue of whether the beatific vision removes freedom, insofar as, traditionally
conceived, it prohibits the choice of sin in the blessed in heaven. Is Christ’s
freedom thus constrained by his beatific knowledge? Provided that there is at
least one account of the freedom of the redeemed whereby they have the beatific
vision, are free, and cannot sin, this objection will fail. Such accounts appear in
the contemporary philosophical literature.70
Second, one might think that freedom requires deliberation, but that someone
with perfect knowledge of what he will do cannot deliberate about his future

68
For more on this point, see Pawl (2019b, chap. 7). For recent discussions of the Eternity Solution,
see Cobreros (2016), Leftow (1991; 2009), Rota (2010), Stump and Kretzmann (1981; 1991),
and Zagzebski (1991). For recent discussions of open theism, see Hasker (1998), Oord, Hasker
and Zimmerman (2011), Rhoda (2007; 2008; 2011), Sanders (1998), and Tuggy (2007).
69
To see this response worked out in more detail, see Pawl (2019b, chap. 7)
70
For more on the freedom of the blessed in heaven, see Brown (2015), Cowan (2011), Henderson
(2014; 2016), Pawl and Timpe (2009; 2013; 2016; 2017), and Tamburro (2014).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 35

actions. No one can take as a deliberative option something he is positive he will


not do. And Christ is positive that he will not do anything but exactly what he
does. So he has no alternative deliberative options. So he cannot deliberate. So
he is not free. Does foreknowledge ruin deliberation and so ruin freedom? In
response, one might deny that deliberation is required for all acts of freedom.
Rather, deliberation is required for cases of ignorance about what one ought to
do. Christ does not fulfill that condition. So he need not deliberate to be free.71
Third, it is impossible for explanation to be symmetric – that is, A can’t explain
B and also B explain A. But, if Christ knew the future, it would be at least possible
for him to look to that future knowledge to determine what to do. That is, he could
reason, “I know I act perfectly always, and so I can look forward to what I will do
to see what I should do in this situation.” If that were even possible, then it would
be possible for his future knowledge to explain his past actions. But his past
actions explain his future knowledge. So it would be possible for explanation to
be symmetric – for his knowledge of his future actions to explain his past acting
(given foreknowledge), and for his past acting to explain his knowledge of his
future actions (since reality grounds truth). Thus, his foreknowledge of his own
future actions implies that something impossible (symmetric explanation) is
possible. That’s contradictory. So he must not have foreknowledge of his own
future actions. In response, one can distinguish types of explanation. True, no two
things can explain each other in the same way. But if Christ’s knowledge of his
future actions is explanatorily prior in a motivational sense to his act, and his
acting is explanatorily prior in a grounding the truth of sense, then we do not have
a vicious explanatory circle here.72

6.3 Death and Descent


Many activities of Christ have not received careful attention in contemporary
philosophy of religion or analytic theology, for instance, his dual nativities, as
Second Constantinople refers to them (from the Father, and from Mary), his
miracles, his resurrection, his ascension, his two energies (as Third Constantinople
defines), etc. Two activities that have received some attention are his death and his
descent into hell. As such, this section focuses on those two events.
How does one explain, ontologically speaking, what happens in Christ’s
death? The standard scholastic answer runs as follows. Christ had both a soul
and a body. At death, his soul and his body separated, but the component parts of
his human nature remained united to the divine nature, though not to each other.
In virtue of that remaining union of the parts of the human nature to the divine

71
For a discussion of the problem of deliberation, see Pawl (2014c; 2019b, chap. 8).
72
For more discussion of the problem of explanatory priority, see Pawl (2014b; 2019b, chap. 8).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
36 Philosophy of Religion

nature, that which the component parts undergo can still rightly be said of
Christ. So, for instance, it is true that he was buried, since his human body was
buried. And it is true that he descended into hell, at least on these medieval
accounts and on the traditional view of the earlier theologians, since his human
soul descended into hell. At the resurrection, those parts of the human nature
were knit together again, and so Christ was an integrated human being again
thereafter.73

6.4 Conclusion
This section has charted the current discussions on the activities of Christ in
analytic philosophy of religion. As noted, various areas remain underdeveloped
in these literatures. The main areas of discussion concern Christ’s knowing,
willing, and death. On the traditional view, he was free with respect to both
wills, though he could not sin. Traditionally, he knew everything with the divine
intellect and all created things with his human intellect, though this did not
negatively affect his freedom or ours. His death has traditionally been explained
as the separation of his body and soul, and his resurrection explained as their
being rejoined.

7 The Fundamental Philosophical Problem


The doctrine of the Incarnation faces an objection of obvious, odious contradic-
tion. For, as found in the writings stemming from the ecumenical councils and in
the work of many of the most influential Christian theologians, it affirms appar-
ently contradictory predicates of Christ. As Richard Cross puts the objection:

[T]he fundamental philosophical problem specific to the doctrine is this: how


is it that one and the same thing could be both divine (and thus, on the face of
it, necessary, and necessarily omniscient, omnipotent, eternal, immutable,
impassible, and impeccable) and human (and thus, on the face of it, have the
complements of all these properties)? (Cross 2011, 453)

Christ is immutable and mutable, impassible and passible, filling all creation
and located at the bosom of Mary, etc. How could one make sense of such
claims?74

73
For current discussion on this issue, see Jaeger (2017), Jaeger and Sienkiewicz (2018), Nevitt
(2016), and Turner (2017; 2019).
74
A brief sampling of other discussions of this problem includes: Adams (2006, 121–123; 2009,
242–243), Arcadi (2018), Davis (2006, 116), Dawson (2004, 161–162), Evans (2006a, 13),
Feenstra (2006, 142–144), Gordon (2016, 64), Gorman (2000a; 2011; 2014; 2016; 2017,
chap. 6), Hebblethwaite (2008, 60), Hick (1989, 415; 2006, 66–70), Hill (2012, 3), Kelly
(1994), Klima (1984), Labooy (2019), Leftow (2011, 316), Le Poidevin (2009b, 704), Loke
(2009, 51; 2011, 493–494), Macquarrie (1990), Moreland and Craig (2003, 597), Morris (1987,
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 37

7.1 A Statement of the Problem


Focusing on just one instance of the problem, we can state it this way:75
1. Anything divine is impassible. (assume)
2. Anything human is passible. (assume)
3. Nothing can be both passible and impassible. (assume)
4. Christ is divine and human. (orthodox Christology)
5. Christ is both passible and impassible. (from 1, 2, 4)
6. Contradiction! (from 3, 5)
We have seen in previous sections that 4 is a requirement of Conciliar
Christology. For just one example, the Nicene Creed states that Jesus Christ is
both God and human; he is “true god” and “became human” (Tanner 1990, 5). In
Section 3, we have seen that the Christology of the ecumenical councils affirms
Christ’s impassibility. We’ve seen in Sections 2 and 4 the need for claiming that
Christ was passible. He suffered at the hands of others, after all. And even if the
evidence in Sections 2, 3, and 4 is insufficient to show that anything human is
passible or anything divine is impassible, the premises themselves are stronger
than we need. We only need the premises that Christ himself is impassible and
passible.
The proponent of orthodox Christology has reason to accept 1, 2, and 4 from
the conciliar texts. Even if there’s some reason to deny 1 or 2, there’s still
independent reason to accept Sub-conclusion 5. For just a brief summary of
reasons from influential theologians and conciliar texts, consider the following.
Athanasius says “that Christ ‘suffered and did not suffer’” (Anatolios 2004, 70).
Aquinas approvingly quotes John of Damascus:

All the properties of the human, just as of the Divine Nature, may be
predicated equally of Christ. Hence Damascene says (De Fide Orth. iii, 4)
that “Christ Who [is] God and Man, is called created and uncreated, passible
and impassible.” (ST III q.16 a.8 ad.2)

Finally, and perhaps most forcefully for those interested in affirming the
orthodox faith concerning the Incarnation, the Fourth Council of
Constantinople says of the seventh ecumenical council:

We also know that the seventh, holy and universal synod, held for the second
time at Nicaea, taught correctly when it professed the one and same Christ as
both invisible and visible lord, incomprehensible and comprehensible,

chap. 1; 2009), Pawl (2014a; 2015; 2016b; 2016d, chaps. 4–7; 2018), Riches (2016, 5, 166),
Senor (2002, 221), Spence (2008, 16), Stump (1989; 2004; 2005, chap. 14), Sturch (1991, chaps.
2, 12), Vallicella (2002), Van Inwagen (1998, secs. 2–4), and Wellum (2016, 446–455).
75
Here I present the argument as I did in Pawl (2015) and summarize the findings from that article
and from chapters 4–7 of Pawl (2016).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
38 Philosophy of Religion

unlimited and limited, incapable and capable of suffering [impassibilem


etiam et passibilem], inexpressible and expressible in writing. (Tanner
1990, 162)

Conciliar Christology requires the affirmation of apparently inconsistent predi-


cates of Christ. That said, not everyone who discusses the Fundamental Problem
feels bound to the conciliar texts. As such, a variety of solutions exists to this
problem, some consistent with the conciliar texts, some not. In the remainder of
this section, I discuss responses to this problem with an eye to which premise
they deny.

7.2 Response 1: Revise Standard Logic


Such a response might go in two ways.76 One might reject the validity of some
inference forms, or one might reject the standard notion of identity, opting
instead for relative identity.

7.2.1 Reject Classical Logic

The rejection of classical logic subdivides, depending on whether the rejected


inference forms are meant to curtail the derivation of a contradiction, or whether
they are meant to stop the spread of contradictions that are accepted. Along the
first divide, one might accept the premises, but reject the universality of some
inference form that allows the conclusion of a contradiction in the first place.
Along the second, one might accept the truth of the contradictory conclusion,
but reject the rules required for the contradiction to spread and ruin everything
(this is sometimes called explosion, with good reason). I discuss each in turn
here.
First, one might accept the claims of the argument but reject some inference
to the contradiction, either the inference to 5 or the final inference to 6. For
instance, Martin Luther (1971, 256) writes in a disputation that “there are
syllogisms that are valid in logic, but not in theology.” Such a move, if viable,
would allow him to avoid the derivation of a contradiction at the cost of
rejecting the universality of some common inference rules (Cf. Dahms
1978, 375).
Second, one might accept the conclusion, but reject the problematic conse-
quences by limiting the entailment relations allowed in the Christological
domain of discourse. Yes, Christ has both complementary predicates apt of
him. What we have here is a contradiction. It is true that a proposition is true and
that very same proposition, at the very same time, in the very same respect, is

76
Let “standard logic” name classical logic plus an absolute identity relation.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 39

false. But logic is such that we cannot derive from that contradiction just
anything. This view rejects the validity of the Law of the Excluded Middle.
Such a response allows the contradiction but precludes the spread of the
contradiction. The work of Jc Beall (2019) is the best place to look for this
strategy.77

7.2.2 Accept Relative Identity

The relative identity strategy contends that some – often all – identity claims are
incomplete without a sortal, a type, under which the two things are identical.
“Clark Kent is Superman” is an incomplete claim it is similar to “Edith is to the
left of.” What is needed in each case is another term: Clark is the same person as
Superman; Edith is to the left of Agnes. Importantly for the relative identity
theorist, two things can be the same x without being the same y. In addition, as
Peter van Inwagen (1998) shows, Leibniz’s Law fails if relative identity is true.
That is, A and B can be the same x and A have a feature without B having that
feature. This is relevant to the Fundamental Problem, since the human-natured
thing can be passible, and the human-natured thing can be the same person as
the divine-natured thing, without the divine-natured thing being passible.
Thus, the inference to 5, which says that one and the same thing is both
passible and impassible, is fallacious, given relative identity. Some people
claim that relative identity is necessary for the doctrine of the Trinity – the
Father and the Son are the same God, but they are not the same person.78 If this
is so, then by accepting the relative identity solution here the Christian does not
incur any additional costs, as they will already have been paid for the doctrine of
the Trinity.79

7.2.3 The Traditional Assumption of Standard Logic


The vast majority of theological authors accept the classical entailment rela-
tions, as we can see from their discussions of their opponents’ views. Very often
the giants of theological discourse argue against their opponents by pointing out
the entailments of their views, then reasoning that, since these entailments are
false, the opponents’ views must be as well. Such argumentation requires the

77
I discuss this strategy and objections to it in more detail in Pawl (2019a). I thank Beall for his
helpful discussion on this strategy.
78
The logic of this solution is difficult. To see it worked out carefully, see Baber (2015; 2016;
2019), Conn (2012), and Jedwab (2015; 2018). To see some objections to relative identity in
a Trinitarian context, see Rea (2003).
79
My thanks to Joseph Jedwab for his patient help in understanding the relative identity response to
the Fundamental Problem.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
40 Philosophy of Religion

use of some inference forms that Luther and Beall (for different reasons) would
want to preclude from the Christological domain.
In what follows, I employ classical logic. I again point the reader to the work
of Beall (2019), though, to see how one could answer the Fundamental Problem
by revising logic (or, perhaps as he would say, by realizing the proper extent of
logic). I also assume an absolute identity relation.

7.3 Response 2: Deny Some Predicates of Christ


On the assumption of classical logic, no one should affirm a contradiction. So,
everyone, friend or foe of Conciliar Christology, will need to deny some
premise of the argument given in Section 7.1. One can deny the relevant
predicates of Christ in at least four ways. One could (1) categorically deny all
the divine predicates, (2) categorically deny all the human predicates, (3)
categorically deny some mixture of predicates, such that one predicate from
each allegedly contradictory pair is denied, or (4) deny some mixture of the
predicates at some time or other, such that one predicate from each pair is denied
for all times, though not necessarily the same predicate at each time.

7.3.1 Deny All Divine Predicates


The proponent of the view under discussion here denies that Christ is divine,
and so denies 4, and is thus not required to affirm 5 or the contradictory 6.
Someone with such a view could affirm that Christ is a mere human,
perhaps a charlatan who deceived his disciples. Or Christ could still be an
important divine messenger, yet a messenger that is not God – at least not God
in the sense meant in traditional Christian doctrine, and so need not have the
divine attributes. In either case, the councils are massively wrong in the
orthodox teaching concerning Christ. One finds this view – that Christ has
a special theological status but is not God – discussed under the title
“Arianism” in the early church. This view, while it resolves the contradiction,
is inconsistent with Conciliar Christology, insofar as Conciliar Christology
requires the divinity of the God-man.80

7.3.2 Deny All Human Predicates

The logic of this maneuver is similar to the logic of the previous response. Such
a thinker denies 4, due to denying the second conjunct, that Christ is human. If
all specifically human predicates are denied, then “human” is denied of Christ,
and so the second conjunct of Premise 4 is false. In such a case, Christ only

80
See Hick (2006) for more on this approach.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 41

appeared to have the human attributes. He only appeared to be passible, for


instance.
Such a view, so far as I know, has no current or even recent defenders in the
philosophical discussions of the Incarnation (in fact, even calling it an
Incarnation is faulty on this view). One finds this view – that Christ only
appeared to become human – discussed under the title “Docetism” (see
Section 4.2.1). This view is inconsistent with Conciliar Christology, insofar as
Conciliar Christology requires the humanity of the God-man.

7.3.3 Deny Some Mixture of Predicates at All Times

This type of response has many advocates today. The idea can be pictured as
follows. One can make a list of all of the pairs of allegedly contradictory
attributes that a God-man would have. Then, one goes through each pair,
crossing off one or the other of the two attributes, such that no pair has both
predicates left at the end of the excising. For instance, perhaps God need not be
impassible. Thus, Premise 1 is false. In another case, perhaps humans need not
really be limited in power, and so the Fundamental Problem constructed with
“omnipotent” and “limited in power” as the allegedly contradictory pair has
a false second premise. And so on. For every pair, at least one of the attributes
fails to be implied by being divine or by being human.
Scholars can use two main methods, always in tandem, to work this strategy.
(To do either alone would be to fall back into the previous two strategies
discussed.) The first is to weaken or eliminate some of the divine attributes
associated with classical theism. On such a view, it is false that God must be
impassible, immutable, eternal, and simple (or false that God is any of these
ways in a robust, problematic sense), and so any form of the Fundamental
Problem that includes those attributes has a false first premise.
The second step is to make a distinction going back at least to St. John of
Damascus (676–749), but more recently revitalized by Thomas Morris. The
Damascene writes:

The whole He, then is perfect God, but not exclusively God, because He is not
only God but also man. Likewise, the whole He is perfect man, but not
exclusively man, because He is not only man but also God.81

Morris (1987, 65–67) makes the same distinction with the terms merely (for
exclusively) and fully (for perfect). Armed with this distinction, one turns to the
Fundamental Problem with an eye toward distinguishing the second premise.
Yes, the responder says, it is true that anything merely human is, say, limited in

81
This quotation is taken, with slight modification, from John of Damascus (2000, 283).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
42 Philosophy of Religion

knowledge or power. But, the responder continues, it is false that anything fully
human is limited in knowledge or power. Similarly, it is true that anything
merely divine is impassible, but false that anything fully divine is impassible. Of
course, it is the fully claim that the objector needs for this argument to apply to
Christ. After all, the merely interpretation of the premise would itself preclude
the truth of Premise 4, that Christ is both God and human – being merely human
implies not being both human and divine. Were the proponent of the
Fundamental Problem to employ the merely premise, the argument’s premises
would contradict one another, and so the argument would be unsound.
To summarize: consider the cases of the Fundamental Problem in which one
means to excise the human predicate. Armed with the distinction between
merely human and fully human, those instances of the Fundamental Problem
can be put in two ways, depending on how one understands the second premise.
On the merely reading, the truth of the second premise contradicts the truth of
the fourth premise, and so the argument is unsound. On the fully reading, this
response claims that the second premise is false, and so the argument is
unsound. Thus, the argument is unsound.
What to make of this strategy? In the end, even if there were a consistent,
uniform manner of applying this strategy in solving the Fundamental
Problem, there’s still the fact that such a move leaves the contradiction in
place for many who affirm the doctrine of the Incarnation. Very many
defenders of the Incarnation take it as a given that the texts cited in
Section 7.1, in which Christ is predicated by apparently incompatible
predicates, as well as many other texts from the councils and theologians,
are to be believed. So, even if there were grounds for denying Premise 1 or
Premise 2 for each presentation of the Fundamental Problem, the traditional
texts preclude simply ruling out the apparently incompatible predications of
Christ. Christ must remain both passible and impassible, for instance, on
this view. Assuming the truth of Premise 3, which this solution does nothing
to challenge, the proponent of traditional Christianity still faces
a contradiction. For this large swath of past and present proponents of the
Incarnation, one must look elsewhere for a reply.
In addition, some varieties of this solution appear to collapse into the
following solution (Section 7.3.4). To see why, consider the Word pre-
Incarnation. He is then both merely and fully divine, and so, on this response,
he is impassible (since “anything merely divine is impassible” is conceded on
this view). Once incarnate, he is fully but not merely divine. Once incarnate, he
is not impassible – were he still impassible, we would not have avoided his
being both passible and impassible in the incarnation, and so this view would
not have resolved the contradiction. Thus, this solution looks to require the
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 43

predications concerning Christ to change their truth value at different times, as


the next solution claims.

7.3.4 Deny Different Predicates at Different Times


Suppose one wanted to accept that Christ had both of the apparently incompat-
ible attributes. How might one go about doing so? One might claim he had them,
but not at the same time or in the same respect. Or, one might deny their
incompatibility full stop, even at the same time in the same respect. This section
canvasses each option, starting in this section with the claim that Christ had the
apparently incompatible predicates, but not at the same time.
The kenotic view takes its lead from a passage of Scripture.82 Paul writes in
Philippians 2:6–8 that Christ:

who, though he was in the form of God, did not regard equality
with God as something to be exploited,
but emptied himself, taking the form of a slave, being born in
human likeness. And being found in human form,
he humbled himself and became obedient to the point of death –
even death on a cross. (NRSV)

This self-emptying – kenosis – is the impetus for the kenotic view. While
kenotic Christology ought not to be seen as merely a response to the Fundamental
Problem, it has been used as one such response, in the following way.
It is true, says the kenotic opponent of the Fundamental Problem, that
anything that is divine is immutable or impassible at some point. And true,
anything human is limited in knowledge and power at some point. But,
importantly, these might not be the same times. And, if not the same times,
then Christ’s being impassible and passible or being omnipotent and limited in
power are no more contradictory than your being seated at one time and
standing at another. What the Fundamental Problem would have to show,
and what its kenotic opponent would fight hard to deny, is that Christ must
have both attributes at the same time. In summary: the first two premises need
an “at some time” modifier added to them, and the last three lines of the
Fundamental Problem need “at the same time” modifiers added to them. Once
the temporal modification is made explicit, the contradiction is no longer
derivable.
This theory faces some objections. First, is there a time at which Christ is both
incarnate and no longer self-emptying? Many have claimed there is. The
doctrine of the exaltation claims that after his resurrection, Christ is glorified
82
For more on kenosis, see Archer (2017), Crisp (2007b, chap. 5), Davis (2011), Evans (2006a),
Feenstra (2006), Le Poidevin (2009b, sec. 6; 2013), Senor (2011), and Thompson (2006).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
44 Philosophy of Religion

to his rightful place, bearing all his divine majesty and attributes. At that point,
Christ is still human, yet he is exalted to his former status. If the doctrine of the
exaltation is true, then kenotic Christology will not provide a universal answer
to the Fundamental Problem. For there will be some time, any post-exaltation
time, at which Christ has all the attributes humans must have in addition to the
attributes of which he supposedly emptied during his earthly mission.83
Second, depending on how one understands kenotic Christology, it can
appear inconsistent with Conciliar Christology.84 For the conciliar texts teach
that Christ, even when incarnate, retained the very divine attributes kenotic
thinkers look to deny of him during his Incarnation. Cyril writes of Christ that,
while a baby, “he filled the whole of creation and was fellow ruler with him who
begot him” (Tanner 1990, 51). And Leo, discussing the very text from
Philippians 2, says that “that self-emptying whereby the invisible rendered
himself visible, and the Creator and Lord of all things chose to join the ranks
of mortals, spelled no failure of power” (Tanner 1990, 78). Christ’s power is not
diminished, according to Leo, even though he becomes a baby. This teaching –
that Christ, even while united completely to the human nature in the Incarnation,
remained with the Father in heaven, ruling all of creation – was given the
pejorative name the extra calvinisticum (roughly, “that extra thing Calvin
added”), in later Protestant polemics.85

7.4 Response 3: “Qua” Responses


The previous section, 7.3.4, focused on a response to the Fundamental Problem
that granted the incompatibility of the predicates in question (e.g., impassible
and passible) when had at the same time and in the same respect, but it denied
that the predicates were apt at the same time. This section focuses on a family of
strategies that accepts that Christ was both, for instance, passible and impass-
ible, at the same time, but denies that he was both in the same respect, and so
denies the actual incompatibility.
Put in terms of the argument presented in Section 7.1, this strategy, as the
previous one, claims that Premise 3 is incompletely stated. It should state that
nothing can be both passible and impassible at the same time, in the same
respect. To derive a contradiction, then, Sub-conclusion 5 would have to say
that Christ is passible and impassible at the same time, in the same respect
(henceforth, I drop the “at the same time” clause, as it is not germane to this

83
For a more in-depth discussion of kenoticism and the exaltation, see Evans (2006b, 200–202),
Feenstra (1989, 144–149), and Pawl (2016d, 114–115).
84
For this argumentation spelled out in more detail, see Pawl (2016d, chap. 5, sect V.b.).
85
For the history of the use of the term, see Willis (1966, 8–25). For more recent work on the extra
calvinisticum, see Gordon (2016) and McGinnis (2014).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 45

response to the argument). The proponent of this solution denies Sub-


conclusion 5 so revised. How?
A common mode of speech, found in the ecumenical councils, says that
Christ is one way according to his human nature and an apparently incompat-
ible way according to his divine nature. For instance, Christ is impassible
according to his divinity and is passible according to his humanity. This
“according to” phrase is often left untranslated as “qua.” So, for instance,
Christ, qua human, is limited in space, but Christ, qua divine, is omnipresent.
Moreover, it is false that Christ, qua human, is omnipresent, and it is false that
Christ, qua divine, is limited in space. As such, the revised Sub-conclusion 5,
which requires that Christ be passible and impassible in the same respect, is
false. What to make of this solution to the Fundamental Problem?
It is a good opening move in response to the Fundamental Problem, but,
spelled out only this far, it is unsatisfactory. An analogy will make the unsatis-
factory nature of the response clear. Suppose that I told you that, against every
prior expectation, I had succeeded in drawing a shape that is both a square and
a circle.86 You would be rightly incredulous. You’d likely say something like the
following:

A necessary condition for being a circle is that the shape have all its points
equidistant from a center point. And a necessary condition for being a square
is that the shape have four 90-degree angles and four straight lines of equal
length. No one shape can fulfill both those necessary conditions. So no one
shape can be both a square and a circle. You must be mistaken.

Now, suppose I responded to you by noting that you’re right concerning the
necessary conditions. Nothing can have those features in the same respect. But
you’ve overlooked that the shape I’ve drawn has all its points equidistant from
a center point according to its circularity, or qua circle, and that it has four
straight lines and four 90-degree angles according to its squareness, or qua
square.
Are all your worries now alleviated? Likely not. In fact, now we have
additional quandaries. What does it mean for one single shape to have incon-
sistent features in these different respects? If we were to look at it, would we
see four angles or not? Without further analysis, this method of solution to
both the Fundamental Problem and the Square Circle Problem (as we might
call it) is incomplete and unhelpful. It is no surprise, then, that contemporary

86
Here I borrow an analogy Hick (1977, 178) makes concerning the coherence of orthodox
Christology. See also the clever work of Angere (2017), though, who argues that there can be
square circles, and, in fact, that it is “quite likely that there are square circles in the universe”
(88). Angere includes a figure, Figure 4, on page 86, which shows a square circle (note: it looks
like a square and not like a circle).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
46 Philosophy of Religion

authors who consider this solution often charge it with “muddying the waters”
(Morris 1987, 49).87
For this reason, it behooves the proponent of this strategy to say more about
what exactly the “qua human” or “qua divine” modifier is doing in the solution.
Elsewhere I’ve distinguished four things the “qua” might be doing, then sub-
divided two of them, for a total of six qua strategies.88 Here I add a seventh
“qua” strategy. There is no space in this Element to provide a complete treat-
ment of these strategies (for that, look at chapter 6 of Pawl (2016d), which itself
is two-thirds the length of the body of this Element). Here we must be content
with a brief summary of the strategies.
Most generally, the strategies can be distinguished by what the “qua” modi-
fies. It could modify the whole of the assertion or a part. If just a part, then, since
there are three parts – “Christ,” “is,” “(im)passible” – there are three ways to
modify a part with the “qua.” It could modify the subject of the assertion
(“Christ-qua-divine”), the copula (“is-qua-divine”), or the predicate
(“[im]passible-qua-divine”).89 The remainder of this section discusses each of
these options. While each option modifies something different, the argumenta-
tive goal is the same. The goal is to revise Premises 3 and 5, claiming that,
understood in the same sense, they are not both true. That is, they all aim to
replace 3 and 5 with the following:

3*. Nothing can be both passible and impassible in the same respect.
5*. Christ is both passible and impassible in the same respect.

3* and 5* are together contradictory. Each qua theory discussed in what follows
provides some way to deny 5*, and so to deny that a contradiction has been
derived.

7.4.1 Modifying the Assertion

This first understanding of the “qua” modifier takes it to modify the whole
assertion: Qua divine, Christ is impassible. The “qua” clause serves to point out
that in virtue of which the subject has the predicate. It is because Christ is divine
that he is impassible.
Scholars have rightly criticized this understanding in the literature. For if
something is a certain way because it is another – if S is P because it is N – then it
follows that S is P. If I am warm-blooded because I am human, then I am warm-

87
See also Van Inwagen (1998, sec. 4) and Holland (2012, 74) for criticisms of the qua move.
88
See Pawl (2015; 2016b; 2016b, chap. 6).
89
For more on divisions of qua clauses, see Adams (2009), Bäck (1997; 2008, 84–87), Cross
(2005, 193–199; 2011, 455–456), Labooy (2019), Morris (1987, 48–49), and Senor (2002,
229–33).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 47

blooded. But then, we haven’t avoided the contradiction between Premises 3


and 5 after all. The qua-modified assertion implies the non-qua-modified asser-
tion with which we started. Those non-qua-modified assertions, though, the
qua-strategist grants, cannot be true of the same thing at the same time. Since, on
this strategy, the qua modification that was intended to get the predicates apt of
Christ in different respects collapses back into the non-qua-modified, problematic
claims, we are no closer to solving the Fundamental Problem.
A novel understanding of modifying the assertion, one that does not fall prey
to the previous objection, has recently been put forward by Beall and Henderson
(2019). Their view takes seriously the prima facie fact that the qua locutions are
sentential operators, modifying the assertion into a different assertion. On this
view, the qua modifiers serve to designate in which “story” about Christ the qua-
modified propositions are true. It is true in the divine story that Christ is
impassible, but true in the human story that he is passible. And, importantly,
stories cannot be combined (they call it “unionized” [2019, 160]) into a single,
larger story. The truth of “Christ is impassible in the divine story” does not, on
this view, imply that “Christ is impassible” full stop; S is P qua N does not imply
S is P. And one cannot derive from “S is P qua N1” and “S is not-P qua N2” that
“S is P and not-P.” Thus, one cannot derive 6. Instead, one derives Christ is
impassible in the divine story and Christ is passible in the human story. That is
not contradictory, according to this solution.
This solution has some costs. It requires that unionizing true stories
about the same individual is not universally valid. It requires that it is not
universally valid to derive a true, qua-unadorned statement about a person
from a true story about that person. And it requires that both “Christ is
impassible” is not true and “Christ is not impassible” is not true (Beall and
Henderson 2019, 166). These considerations are not news to the authors,
who carefully consider nine objections to the view in their article, includ-
ing some of these objections.

7.4.2 Modifying the Subject


The second understanding of the “qua” modifier takes it to modify the subject:
Christ-qua-divine is impassible. The “qua” serves to disambiguate the subject of
the predicate. It is Christ-qua-divine that is impassible; it is Christ-qua-human
that is passible. Which premise or inference of the Fundamental Problem would
the proponent of this strategy challenge? Again, it would revise Premise 3 to
include an in the same respect clause, and it would say that Premise 5 is false –
Christ, the one person, is not both passible and impassible in the same respect.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
48 Philosophy of Religion

The thing, Christ-qua-human, is passible; the thing, Christ-qua-divine, is


impassible.
To see the problems with this view, consider the relations between the
subjects of the predications so far discussed: Christ, Christ-qua-divine, and
Christ-qua-human. Are any of these three identical with one another? Either
they are identical or they are not. What follows is a proof by cases.
Suppose that they all name the same entity. At least two problems arise. First,
if a and b are identical, anything true of a is true of b and vice versa.90 So, if
Christ-qua-divine is identical to Christ-qua-human, then, since Christ-qua-
divine is impassible, it follows that Christ-qua-human is impassible. But
that’s false – Christ-qua-human is not impassible.
Second, if they are all identical (in the nonrelativistic sense), then this
solution does not solve the contradiction after all. This solution is premised
on the acceptance of the two attributes being incompatible when said of the
same thing at the same time in the same respect. And it attempted to resolve the
Fundamental Problem by noting that these predicates were not said in the same
respect, since, really, they were not said of the same thing. But then, if the next
move were to identify the subjects, subjects differentiated to avoid contradic-
tion, then we haven’t made any headway in solving the Fundamental Problem.
I think the strategy must therefore require some real distinction between at least
two of the subjects discussed.
Suppose, then, for the other case in our proof by cases, that there are at least
two separate entities among the three. In such a case, perhaps one could stop
both allegedly contradictory predicates from spreading to one subject. Here too
the view faces problems.
First, it seems contrary to the Communication of Idioms (see Section 5.2).
Christ’s human nature exists in a manner that would typically make a person
with a human nature be passible. And his divine nature exists in a manner that
would typically make a person with that divine nature be impassible. The
Communication of Idioms would imply that the one person is both passible
and impassible, then. But this view must deny that implication on pain of
reinstating the contradiction.
Second, this view sits poorly with Conciliar Christology, which requires both
the divine and human attributes to be said of one and the same person. As just
one instance of many, Cyril says in his third letter to Nestorius, “in thinking
rightly, we refer both the human and divine expressions to the same person”
(Tanner 1990, 55). If this qua strategy were to allow the predication of both

90
The conditional in this sentence is sometimes referred to as Leibniz’s Law. To see a response that
rejects the use of Leibniz’s Law, see the relative identity response in Section 7.2.2.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 49

expressions to the one person, it would have no means by which to claim that the
(allegedly) contradictory predicates do not imply a contradiction.91

7.4.3 Modifying the Copula


The third understanding of the “qua” modifier takes it to modify the copula of
the assertion: Christ is-qua-divine impassible, or, put otherwise, Christ is-
divinely impassible.92 The “qua” serves to disambiguate the different copulas
involved in the assertions. The opponent of the Fundamental Problem who
prefers this strategy will claim that while it is true that the predicates cannot be
apt of Christ at the same time in the same respect, it is false that they are apt of
Christ in the same respect, and so Premise 5, when revised as 5*, is false. Christ
is-divinely impassible and the very same Christ is-humanly passible.
This strategy for answering the Fundamental Problem can be subdivided in
two ways. It could be that there are different copulas – is-divinely, is-humanly,
is-bovinely, etc. – for each kind of thing that there is. Or, it could be that there is
only a single copula with a variable built into it – is-xly – and the difference
between the Christological claims comes down to what one substitutes in for x.
Since the latter includes a variable for which one can substitute in nature types,
we can call that the substitutional copula view. Since the former includes no
variables, but rather a plethora of different copulas, we can call that the non-
substitutional copula view.93 Consider each.
The non-substitutional copula view requires multiple copulas, minimally,
is-divinely and is-humanly. Limiting the number of copulas to two, though,
would imply that we couldn’t say anything about anything aside from humans
and the divine. “Fido is a dog” is true, but that “is” cannot be the divine or
human copula. It seems, then, that for any type of thing, there must be
a distinct copula for it. Is there, in addition to these narrow copulas,
a universal copula as well? The non-substitutional theorist should deny the
existence of a universal copula. For, were there such a thing, an inference from
“Christ is-divinely impassible” to “Christ is impassible” would seem war-
ranted. If it weren’t, in what sense would the copula be universal? If one
couldn’t go from the particular to the universal copula, then, since everything
is of some type, there would be no instance in which one could employ
a universal copula. Far from being universal, then, it would be inapplicable.
The substitutional copula view does have a universal copula, unlike the non-
substitutional view.

91
For a recent defense of the strategy of this section, see Labooy (2019).
92
I discuss this qua strategy at length in Pawl (2016d, 143–150).
93
I discuss these views in depth in Pawl (2016d, 143–150).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
50 Philosophy of Religion

Both the substitutional view and the non-substitutional view, were they
employed, have a shared cost. Each would require a modification in our
standard logical formulation of arguments. For standard formulation of logic
has no means of demarcating various copulas.
Additional adverbial solutions to the Fundamental Problem are rare in the
contemporary discussion.94 One instance of a different sort of adverbial solu-
tion comes from Bohn (2012), who distinguishes, not between humanly and
divinely had predicates, but between merely, fundamentally, and essentially
true propositions.

7.4.4 Modifying the Predicate

The fourth understanding of the “qua” modifier takes it to modify the


predicate of the assertion: Christ is impassible-qua-divine.95 The “qua”
serves to disambiguate the different predicates involved in the assertions.
The opponent of the Fundamental Problem who prefers this strategy will
claim that while it is true that “passible” and “impassible” cannot be apt of
Christ at the same time in the same respect, such is not what her Christology
requires. Christ isn’t both “passible” and “impassible”; Christ is “passible-
qua-human” and “impassible-qua-divine.” While “passible” and “impass-
ible” are contradictories on this solution to the Fundamental Problem, there’s
no good reason to think that “passible-qua-human” and “impassible-qua-
divine” are. Thus, this view of the predicates required of Christ does not
predicate allegedly incompatible predicates of Christ. Again, Premise 5* of
the Fundamental Problem is false.
This final “qua” strategy subdivides between a substitutional and a non-
substitutional conception of the qua-relativized predicates. Are there various
predicates – passible-qua-human, passible-qua-dog, passible-qua-cat, etc.? Or
is there just one relational predicate – passible-qua-x, where one can substitute
any kind term for x? Each version of the strategy has its own benefits and costs.
The non-substitutional qua modifying of the predicates yields some counter-
intuitive results. For instance, the word “animal” cannot be used univocally of
both dogs and humans. For humans are really animals-qua-human, and dogs are
really animals-qua-canine. And no human is anything qua-canine, neither is any
dog anything qua-human. So there is no shared term to use here for both dogs
and humans. That’s surprising.

94
Distinctions in ways of being are less rare, however. See, for instance, McDaniel (2017).
95
I discuss this version of the qua strategy at length in Pawl (2016d, 129–143). See Gorman (2014;
2016; 2017, chap. 6) and Labooy (2019) for a recent sympathetic discussion of these “qua”
clause strategies.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 51

The proponent of the view under discussion here should fight the urge to
reason as follows: “I am an animal-qua-human, so I am an animal full stop.
Similar reasoning shows that Fido is an animal full stop. So, there is a shared
term.” Such reasoning – from the qua-modified predicate to the qua-unadorned
predicate – would cause problems elsewhere. For, if generalized, we could do
the same with the predicates correctly said of Christ, in which case we are back
to passible and impassible being said of Christ full stop in the same respect.
A second problem the non-substitutional qua modifying of predicates faces
are predicates that are not apt of Christ in virtue of either nature alone, as
discussed in Section 5.2. For instance, Christ is “two-natured.” But he is not
two-natured-qua-human, and he isn’t two-natured-qua-divine. How does one
explain the qua modification of predicates here?
One thing that could be done is a restricting of which predicates get
qua modified. Maybe animal and two-natured don’t get modified. One hopes
that the reasoning in determining which to modify is something more than
merely cataloging the ones that cause problems, then labeling those as the
modified ones. One would want a principle here, to safeguard against it being
ad hoc.
The substitutional qua modifying of the predicates fares better with univocal
predication across mundane cases. For both I and the dog are animal-qua-
something. That something is different in each case, but the same predicate is
apt of each of us. Similarly, parent-of-someone can be said of both me and many
of my readers, even if it isn’t the same someone in all but one case (that is,
supposing that my wife ever reads this Element). The substitutionary qua modi-
fication of the predicates will still need some means by which to claim that
Christ is two-natured, but not two-natured due to either nature alone. It will still
require some principle for determining which predicates get modified and
which do not.

7.5 Response 4: Revise the Truth Conditions


of the Predicates in Question
Consider a final response. It accepts standard logic and the claim that Christ is,
at the same time, and in the same respects, both passible and impassible, unlike
all the other previous responses. It takes as its target Premise 3 (or 3*) of the
argument: that nothing can be both passible and impassible at the same time, in
the same respect.
If we understand passibility and impassibility as related to each other as
logical complements – P and ~P – then it is clear that nothing can be both
passible and impassible. For to be both would be to imply the truth of

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
52 Philosophy of Religion

a contradiction, and we are supposing here that no contradictions are true. But
what if they were related, not as contradictories, but as subcontraries? That is to
say, what if they can both be apt of a thing, but it is false that they can both be
inapt of a thing? I’ve presented the logic in detail elsewhere (Pawl 2014a;
2016d, chap. 7). Here I summarize it.
To see the distinction between contradictories and subcontraries, consider
where the negation goes in the analysis of the predicate “impassible.” Is to be
impassible not to be able to be causally affected, with the negation on the
outside, at the beginning of the analysis? Or does the negation go on the inside,
for instance, by saying that to be impassible is to have a nature that is not able
to be causally affected?96 Understood in this second way, provided that one
person has two natures, where one such nature can be causally affected and the
other cannot be causally affected, then that one person can fulfill the condi-
tions required for being both passible and impassible at the same time.
Provided that everything has at least one nature (Christ alone, on traditional
Christianity, has more than one), that nature will be either causally affectable
or not. So everything will be either passible or impassible, and it is possible for
one thing – a two-natured person – to be both. In other words, the predicates
“passible” and “impassible” will be related as subcontraries, as defined in the
previous paragraph. Is Christ passible? Yes; he has a nature that can be
causally affected, and that’s what’s required to be passible. Is Christ impass-
ible? Yes; he has a nature that cannot be causally affected, and that’s what’s
required to be impassible. Both terms are truthfully predicated of the one
person Jesus Christ.
In what sense are these terms said in the same respect of Christ? Well,
“passible” is said in exactly the same way it is said of me. And “impass-
ible” is said in exactly the same way it is said of the Holy Spirit. And
neither I nor the Holy Spirit is both impassible and passible, owing to the
fact that each of us has but one nature, and no one-natured thing could
fulfill the conditions required for being both passible and impassible.
Those two predicates would imply a contradiction were they both said of
me. Given that they are contradictory when said of one-natured me, they
must be said in the same respect of me. But they are said in those very
same respects of Christ as they would be of me. So they are said in the
same respect of Christ as well.
This view too faces some objections. It requires us to understand natures such
that they can be causally affected. Such a notion of natures is contested. That

96
This is an old distinction. Gabriel Biel (died 1495) used exactly this strategy to solve the
Fundamental Problem; see Pawl (2019a, 442–447).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
The Incarnation 53

said, one finds such an understanding in the councils, where, as we saw in


Section 4.1.3, Christ’s human nature is said to hang on a cross.97

7.6 Conclusion
This section has discussed the Fundamental Problem of Christology – the
question of how one thing could be both God and human, since, as God, he
must have certain features, and, as human, have apparently contradictory
features. For ease of explication, the section focused on one pair of features –
impassibility and passibility – though the intent was to generalize to other
features. The section canvassed four types of response, many subdivided
further.
First, one might revise standard logic. This can be done by swapping absolute
identity for relative identity. Or it can be done by rejecting the universal
applicability of logical inference (as Luther does) or by accepting universal
applicability but rejecting the validity of some classical inference forms, as
Beall does.
Second, one might deny some of the worrisome predicates of Christ. Perhaps
he isn’t really impassible. Or perhaps he isn’t really ignorant. Or perhaps he has
both allegedly contradictory predicates apt of him, but he doesn’t have them apt
of him at the same time. In any case, this suite of strategies takes its defining
feature to be denying some predicate of Christ, either at a time or categorically.
Third, one might accept standard logic, accept that the worrisome pairs of
predicates are both apt (true) of Christ at the same time, but deny that they are
true in the same respect. This move is typically made with the use of “qua”
clauses. One must ask: how does the “qua” work? I provided four ways of
understanding it. The “qua” clause could modify the whole assertion, modify
the subject of the assertion, modify the copula, or modify the predicate. The first
and last two methods further subdivide.
Fourth and finally, one might accept standard logic, accept the worrisome
pairs of predicates are true of Christ at the same time, in the same respect, but
deny that the worrisome predicate pairs are inconsistent. To do this, one could
revise one’s understanding of the predicate pairs so that they are subcontraries,
each apt of a thing in virtue of some nature it has. Given that Christ has two
natures, he (and only he, on traditional Christian doctrine) is in a position to
satisfy both apparently contradictory predicates. Those predicates are inconsist-
ent for all one-natured other things.

97
For more on this robust notion of the human nature, see Pawl (2016d, 227–231; 2019b, 29–30;
2020).
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
54 Philosophy of Religion

Each method has its own costs and benefits, which I have only gestured
toward here, while providing references to fuller discussions in the footnotes.

8 Conclusion
This Element has canvassed the current philosophical work on the doctrine of
the Incarnation, the Christian view that God the Son assumed a human nature,
and so became human. It focused on the metaphysical aspects of the traditional
doctrine as espoused in the ecumenical councils of the first eight centuries of the
Christian community, prior to the Great Schism.
The goal of this Element has been to present the content of Conciliar
Christology, the main philosophical objection to that doctrine, and the most
common responses to that objection. My hope is that the reader can use this
information, along with the copious footnotes, to investigate the doctrine of the
Incarnation more deeply.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
References
Adams, Marilyn McCord. 1985. “The Metaphysics of the Incarnation in Some
Fourteenth-Century Franciscans.” In Essays Honoring Allan B. Wolter, edited
by William A. Frank and Girard J. Etzkorn, 21–57. New York: Franciscan
Institute.
1999. What Sort of Human Nature? Medieval Philosophy and the Systematics
of Christology. Milwaukee, WI: Marquette University Press.
2000. Horrendous Evils and the Goodness of God. Ithaca, NY: Cornell
University Press.
2004. “The Metaphysical Size Gap.” Sewanee Theological Review 47 (2):
129–144.
2005. “What’s Metaphysically Special about Supposits? Some Medieval
Variations on Aristotelian Substance.” Aristotelian Society
Supplementary Volume 79 (1): 15–52.
2006. Christ and Horrors: The Coherence of Christology. 1st edition.
Cambridge: Cambridge University Press.
2009. “Christ As God-Man, Metaphysically Construed.” In Oxford Readings
in Philosophical Theology, edited by Michael C. Rea, 239–263. Oxford:
Oxford University Press.
de Aldama, SJ, Joseph A., and Iesu Solano, SJ. 2014. Sacrae Theologiae Summa
IIIA: On the Incarnate Word • On the Blessed Virgin Mary. Translated by
Kenneth Baker, SJ. 1st edition. Saddle River, NJ: Keep the Faith.
Alfeyev, Metropolitan Hilarion. 2012. Orthodox Christianity: Doctrine and
Teaching of the Orthodox Church. Yonkers, NY: St. Vladimirs Seminary Press.
Anatolios, Khaled. 2004. Athanasius. New York: Routledge.
Angere, Staffan. 2017. “The Square Circle.” Metaphilosophy 48 (1–2): 79–95.
https://doi.org/10.1111/meta.12224.
Aquinas, Thomas. 1954. The Disputed Questions on Truth (in Three Volumes).
Translated by Robert Schmidt. Chicago: Regnery
2012. Commentary on the Letter of Saint Paul to the Romans. Translated by
Fabian R. Larcher, John Mortensen, and Enrique Alarcón. Lander, WY:
Aquinas Institute for the Study of Sacred Doctrine.
2013. Commentary on the Gospel of Matthew 1–12. Lander, WY: Aquinas
Institute for the Study of Sacred Doctrine.
Arcadi, James M. 2018. “Recent Developments in Analytic Christology.”
Philosophy Compass 0 (0): e12480. https://doi.org/10.1111/phc3
.12480.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
56 References

Archer, Joel. 2017. “Kenosis, Omniscience, and the Anselmian Concept of


Divinity.” Religious Studies, March, 1–13. https://doi.org/10.1017
/S0034412517000051.
Arendzen, J. P. 1941. Whom Do You Say-?: A Study in the Doctrine of the
Incarnation. New York: Sheed and Ward.
Baber, H. E. 2015. “The Trinity.” Faith and Philosophy 32 (2): 161–171. https://
doi.org/10.5840/faithphil201541336.
2016. “Trinity, Generality, and Dominance.” Religious Studies 52 (4):
435–449. https://doi.org/10.1017/S003441251500058X.
2019. The Trinity: A Philosophical Investigation. London: Scm Press.
Bäck, Allan T. 1997. On Reduplication: Logical Theories of Qualification
(Studien Und Texte Zur Geistesgeschichte Des Mittelalters, No 49).
Leiden: Brill.
2008. “Aquinas on the Incarnation.” The New Scholasticism 56 (2): 127–145.
Baker, Kenneth. 2013. Jesus Christ – True God and True Man: A Handbook on
Christology for Non-theologians. South Bend, IN: Saint Augustine’s Press.
Bartel, T. W. 1991. “Like Us in All Things, Apart from Sin?” Journal of
Philosophical Research 16: 19–52.
1995. “Why the Philosophical Problems of Chalcedonian Christology Have
Not Gone Away.” Heythrop Journal 36 (2): 153–172.
Bavinck, Herman. 2006. Reformed Dogmatics, Vol. 3: Sin and Salvation in
Christ. Grand Rapids, MI: Baker Academic.
Beall, Jc. 2019. “Christ – A Contradiction: A Defense of Contradictory
Christology.” Journal of Analytic Theology 7: 400–433.
Beall, Jc, and Jared Henderson. 2019. “A Neglected Qua Solution to the
Fundamental Problem of Christology.” Faith and Philosophy. May 16,
2019. https://doi.org/10.5840/faithphil201957124.
Bird, Michael F., Dr. Craig A. Evans, Simon Gathercole, Charles E. Hill, and
Chris Tilling. 2014. How God Became Jesus: The Real Origins of Belief in
Jesus’ Divine Nature: A Response to Bart D. Ehrman. 1st edition. Grand
Rapids, MI: Zondervan Academic.
Bohn, Einar Duenger. 2012. “The Logic of the Incarnation.” In Logic in
Orthodox Christian Thinking, edited by Andrew Schumann, 104–121.
Lancaster, UK: Gazelle Distribution.
Bonting, Sjoerd L. 2003. “Theological Implications of Possible Extraterrestrial
Life.” Zygon 38 (3): 587–602. https://doi.org/10.1111/1467–9744.00523.
Brazier, Paul. 2013. “C. S. Lewis: The Question of Multiple Incarnations.”
Heythrop Journal 55: 391–408. https://doi.org/10.1111/heyj.12049.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
References 57

Brown, Christopher M. 2015. “Making the Best Even Better: Modifying


Pawl and Timpe’s Solution to the Problem of Heavenly Freedom.”
Faith and Philosophy 32 (1): 63–80.
Brümmer, Vincent. 1984. “Divine Impeccability.” Religious Studies 20 (2):
203–214. https://doi.org/10.2307/20006044.
Canham, Michael McGhee. 2000. “Potuit Non Peccare Or Non Potuit Peccare:
Evangelicals, Hermeneutics, and the Impeccability Debate.” The Masters
Seminary Journal 11 (1): 93–114
Carlson, John W. 2012. Words of Wisdom: A Philosophical Dictionary for the
Perennial Tradition. 1st edition. South Bend, IN: University of Notre Dame
Press.
Carter, W. R. 1985. “Impeccability Revisited.” Analysis 45 (1): 52–55. https://
doi.org/10.2307/3327405.
Cobreros, Pablo. 2016. “Supervaluationism and the Timeless Solution to the
Foreknowledge Problem.” Scientia et Fides 4 (1): 61–75.
Conn, Christopher Hughes. 2012. “Relative Identity, Singular Reference, and
the Incarnation: A Response to Le Poidevin.” Religious Studies 48 (01):
61–82. https://doi.org/10.1017/S0034412511000035.
Couehoven, Jesse. 2012. “The Necessities of Perfected Freedom.” International
Journal of Systematic Theology 14 (4): 396–419.
Cowan, Steven B. 2011. “Compatibilism and the Sinlessness of the Redeemed
in Heaven.” Faith and Philosophy 28 (4): 416–431.
Craig, William Lane. 2006. “Flint’s Radical Molinist Christology Not Radical
Enough.” Faith and Philosophy: Journal of the Society of Christian
Philosophers 23 (1): 55–64.
2018. The Atonement. Cambri: Cambridge University Press.
Crisp, Oliver D. 2004. “Did Christ Have a Fallen Human Nature?”
International Journal of Systematic Theology 6 (3): 270–88.
2007a. “William Shedd on Christ’s Impeccability.” Philosophia Christi 9 (1):
165–188.
2007b. Divinity and Humanity: The Incarnation Reconsidered. 1st edition.
Cambridge: Cambridge University Press.
2007c. “Was Christ Sinless or Impeccable?” Irish Theological Quarterly 72
(2): 168–186. https://doi.org/10.1177/0021140007082165.
2008. “Multiple Incarnations.” In Reason, Faith and History: Philosophical
Essays for Paul Helm, edited by Martin Stone, 219–238. Burlington, VT:
Ashgate.
2009. God Incarnate: Explorations in Christology. 1st edition. New York:
T&T Clark.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
58 References

2011. “Compositional Christology without Nestorianism.” In The


Metaphysics of the Incarnation, edited by Anna Marmodoro and
Jonathan Hill, 45–66. Oxford: Oxford University Press, USA.
2016. The Word Enfleshed: Exploring the Person and Work of Christ. Grand
Rapids, MI: Baker Academic.
Cross, Richard. 2005. The Metaphysics of the Incarnation: Thomas Aquinas to
Duns Scotus. Oxford: Oxford University Press.
2011. “The Incarnation.” In The Oxford Handbook of Philosophical
Theology, edited by Thomas P. Flint and Michael Rea, 452–475. Oxford:
Oxford University Press.
2019. Communicatio Idiomatum: Reformation Christological Debates.
Oxford: Oxford University Press.
Cuff, Andrew Jacob. 2015. “Duns Scotus and Jacques de Thérines on Free Will
and the Word’s Assumption of Human Nature.” Philosophy and Theology
27 (2): 351–389.
Cupitt, Don. 1977. “The Christ of Christendom.” In The Myth of God Incarnate,
edited by John Hick, 133–147. Philadelphia, PA: Westminster Press.
Dahms, John V. 1978. “How Reliable Is Logic?” Journal of the Evangelical
Theological Society 21 (4): 369–380.
Dalmau, Joseph. 2016. On the One and Triune God. Translated by Kenneth
Baker. Vol. IIA. Sacrae Theologiae Summa IIA. Saddle River, NJ: Keep
the Faith.
Davidson, Ivor J. 2008. “Pondering the Sinlessness of Jesus Christ: Moral
Christologies and the Witness of Scripture.” International Journal of
Systematic Theology 10 (4): 372–398. https://doi.org/10.1111/j.1468–
2400.2007.00331.x.
Davies, Brian. 2004. An Introduction to the Philosophy of Religion. 3rd edition.
Oxford: Oxford University Press.
Davies, Paul. 2003. “ET and God.” Atlantic Monthly 292 (2): 112–118.
Davis, Stephen T. 2006. “Is Kenosis Orthodox?” In Exploring Kenotic
Christology, edited by C. Stephen Evans, 112–138. Oxford: Oxford
University Press.
2011. “The Metaphysics of Kenosis.” In The Metaphysics of the Incarnation,
edited by Anna Marmodoro and Jonathan Hill, 114–133. Oxford: Oxford
University Press.
Dawson, Samuel. 2004. “Is There a Contradiction in the Person of Christ? The
Importance of the Dual Nature and Dual Consciousness of Jesus Christ.”
Detroit Baptist Seminary Journal 9: 161–181.
Deng, Natalja. 2018. God and Time. Cambridge: Cambridge University
Press.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
References 59

Denzinger, Henry. 2002. The Sources of Catholic Dogma. Fitzwilliam, NH:


Loreto.
Deweese, Garrett. 2007. “One Person, Two Natures: Two Metaphysical Models
of the Incarnation.” In Jesus in Trinitiarian Perspective, edited by
Fred Sanders and Klaus Issler, 114–153. Nashville, TN: B&H.
Dorner, Isaak August. 1994. Divine Immutability: A Critical Reconsideration.
1st edition. Minneapolis, MN: Fortress Press.
Dubray, Charles. 1911. “Nature.” In The Catholic Encyclopedia. Vol. 10.
New York: Robert Appleton Company. www.newadvent.org/cathen/
10715a.htm.
Ehrman, Bart D. 2013. Did Jesus Exist? The Historical Argument for Jesus of
Nazareth. New York: HarperOne.
Erickson, Millard J. 1996. The Word Became Flesh: A Contemporary
Incarnational Christology. MIGrand Rapids,: Baker Academic.
Evans, C. Stephen. 2006a. Exploring Kenotic Christology: The Self-Emptying
of God. Oxford: Oxford University Press.
2006b. “Kenotic Christology and the Nature of God.” In Exploring Kenotic
Christology, edited by C. Stephen Evans, 190–217. Oxford: Oxford
University Press.
Feenstra, Ronald. 1989. “Reconsidering Kenotic Christology.” In Trinity,
Incarnation and Atonement, edited by Ronald Feenstra and
Cornelius Plantinga, 128–152. Notre Dame, IN: University of Notre Dame
Press.
2006. “A Kenotic Christology of Divine Attributes.” In Exploring Kenotic
Christology, edited by C. Stephen Evans, 139–164. Oxford: Oxford
University Press.
Ferrier, Francis. 1962. What Is the Incarnation? 1st edition. New York:
Hawthorn Books.
Fisher, Christopher L., and David Fergusson. 2006. “Karl Rahner and The
Extra-Terrestrial Intelligence Question.” Heythrop Journal 47 (2):
275–290.
Fisk, Philip J. 2007. “Jonathan Edwards’s Freedom of the Will and His Defence
of the Impeccability of Jesus Christ.” Scottish Journal of Theology 60 (03):
309–325. https://doi.org/10.1017/S0036930607003304.
Flint, Thomas P. 2001. “The Possibilities of Incarnation: Some Radical Molinist
Suggestions.” Religious Studies 37 (3): 307–320.
2011. “Should Concretists Part with Mereological Models of the
Incarnation?” In The Metaphysics of the Incarnation, edited by
Anna Marmodoro and Jonathan Hill, 67–87. Oxford: Oxford University
Press.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
60 References

2012. “Molinism and Incarnation.” In Molinism: The Contemporary


Debate, edited by Ken Perszyk, 187–207. Oxford: Oxford University
Press.
2015. “Is Model T Rattle-Free?” Faith and Philosophy 32 (2): 177–181.
Forrest, Peter. 2009. “The Incarnation: A Philosophical Case for Kenosis.” In
Oxford Readings in Philosophical Theology, edited by Michael C. Rea,
225–238. Oxford: Oxford University Press.
Freddoso, Alfred. 1983. “Logic, Ontology and Ockham’s Christology.” New
Scholasticism 57 (3): 293–330.
1986. “Human Nature, Potency and the Incarnation.” Faith and Philosophy 3
(1): 27–53.
Funkhouser, Eric. 2006. “On Privileging God’s Moral Goodness.” Faith and
Philosophy 23 (4): 409–422.
Gaine, Simon Francis. 2015a. “Christ’s Acquired Knowledge According to
Thomas Aquinas: How Aquinas’s Philosophy Helped and Hindered His
Account.” New Blackfriars 96 (1063): 255–268.
2015b. Did the Saviour See the Father? London; Bloomsbury T&T
Clark.
Garcia, Laura L. 1987. “The Essential Moral Perfection of God.” Religious
Studies 23 (1): 137–144.
Geddes, Leonard. 1911. “Person.” In The Catholic Encyclopedia. Vol. 11.
New York: Robert Appleton. www.newadvent.org/cathen/11726a.htm.
Gellman, Jerome. 1977. “Omnipotence and Impeccability.” New Scholasticism
51 (December): 21–37.
George, Marie I. 2001. “Aquinas on Intelligent Extra-Terrestrial Life.” The
Thomist 65 (2): 239–258.
Gondreau, Paul. 2009. “St. Thomas Aquinas, the Communication of Idioms,
and the Suffering of Christ in the Garden of Gethsemane.” In Divine
Impassibility and the Mystery of Human Suffering, edited by James
F. Keating and Thomas Joseph White O.P. , 214–245. Grand Rapids, MI:
Eerdmans.
2018. The Passions of Christ’s Soul in the Theology of St. Thomas Aquinas.
Cluny Media.
Gordon, James R. 2016. The Holy One in Our Midst: An Essay on the Flesh of
Christ. Minneapolis, MN: Fortress Press.
Gorman, Michael. 2000a. “Personal Unity and the Problem of Christ’s
Knowledge.” Proceedings of the American Catholic Philosophical
Association 74: 175–186.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
References 61

2000b. “Uses of the Person-Nature Distinction in Thomas’s


Christology.” Recherches de Theologie et Philosophie Medievales 67
(1): 58–79.
2011. “Incarnation.” In The Oxford Handbook of Aquinas, edited by
Brian Davies and Eleonore Stump, 428–435. Oxford: Oxford University
Press.
2014. “Christological Consistency and the Reduplicative Qua.” Journal of
Analytic Theology 2 (1): 86–100. https://doi.org/10.12978/jat.2014–1
.120811061413a.
2016. “Classical Theism, Classical Anthropology, and the Christological
Coherence Problem.” Faith and Philosophy 33 (3): 278–292. https://doi
.org/10.5840/faithphil201662164.
2017. Aquinas on the Metaphysics of the Hypostatic Union. Cambridge:
Cambridge University Press.
Green, Adam. 2017. “Omnisubjectivity and Incarnation.” Topoi 36 (4):
693–701.
Harris, Kevin, and William Lane Craig. n.d. “Does Dr. Craig Have an Orthodox
Christology?” Reasonable Faith. www.reasonablefaith.org/media/reason
able-faith-podcast/does-dr.-craig-have-an-orthodox-christology.
Hart, Trevor A. 1995. “Sinlessness and Moral Responsibility: A Problem in
Christology.” Scottish Journal of Theology 48 (01): 37–54.
Hasker, William. 1998. God, Time, and Knowledge. Ithaca, NY: Cornell Univ
Press.
2015. “Getting That Model T Back On the Road.” Faith and Philosophy 32
(2): 172–176.
2016. “A Compositional Incarnation.” Religious Studies. December 2016.
2017. Incarnation: The Avatar Model. Oxford: Oxford University Press.
www.oxfordscholarship.com/view/10.1093/oso/9780198806967.001
.0001/oso-9780198806967-chapter-6.
2019. “The One Divine Nature.” TheoLogica: An International Journal for
Philosophy of Religion and Philosophical Theology 3 (2). https://doi.org
/10.14428/thl.v3i1.2893.
Hebblethwaite, Brian. 2001. “The Impossibility of Multiple Incarnations.”
Theology 104 (821): 323–334. https://doi.org/10.1177/0040571X
0110400502.
2008. Philosophical Theology and Christian Doctrine. Oxford: Wiley.
Henderson, Luke. 2014. “Character-Development and Heaven.” International
Journal for Philosophy of Religion 76 (3): 319–330.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
62 References

2016. “Impeccability and Perfect Virtue.” Religious Studies, September,


1–20. https://doi.org/10.1017/S003441251600024X.
Hick, John. 1977. The Myth of God Incarnate. Louisville,KY: Westminster
Press.
1989. “The Logic of God Incarnate.” Religious Studies 25 (4): 409–423.
2006. The Metaphor of God Incarnate, Second Edition: Christology in
a Pluralistic Age. Louisville, KY: Westminster John Knox Press.
Hill, Jonathan. 2012. “Incarnation, Timelessness, and Exaltation.” Faith and
Philosophy 29 (1): 3–29.
Hipp, Stephen. 2001. “Person” in Christian Tradition and in the Conception of
Saint Albert the Great. Münster: Aschendorff.
Hofmann, Wilhelm, Roy F. Baumeister, Georg Förster, and Kathleen D. Vohs.
2012. “Everyday Temptations: An Experience Sampling Study of Desire,
Conflict, and Self-Control.” Journal of Personality and Social Psychology
102 (6): 1318–1335. https://doi.org/10.1037/a0026545.
Holland, Richard A. 2012. God, Time, and the Incarnation. Eugene,OR: Wipf &
Stock.
Jaeger, Andrew. 2017. “Hylemorphic Animalism and the Incarnational Problem
of Identity.” Journal of Analytic Theology 5 (1): 145–162.
Jaeger, Andrew J., and Jeremy Sienkiewicz. 2018. “Matter without Form: The
Ontological Status of Christ’s Dead Body.” Journal of Analytic Theology 6
(1): 131–145.
Jedwab, Joseph. 2011. “The Incarnation and Unity of Consciousness.” In The
Metaphysics of the Incarnation, edited by Anna Marmodoro and
Jonathan Hill, 168–185. Oxford: Oxford University Press.
2015. “Against the Geachian Theory of the Trinity and Incarnation.” Faith
and Philosophy 32 (2): 125–145.
2018. “Timothy Pawl. In Defense of Conciliar Christology.” Journal of
Analytic Theology 6 (1): 743–747.
John of Damascus. 1958. Fathers of the Church: Saint John of Damascus :
Writings. Translated by Frederic Chase. Washington, DC: Catholic
University of America Press.
. 2000. Saint John of Damascus: Writings. Translated by Frederic Chase.
Washington, DC: Catholic University of Amer Press.
Kavanagh, David J., Jackie Andrade, and Jon May. 2005. “Imaginary Relish
and Exquisite Torture: The Elaborated Intrusion Theory of Desire.”
Psychological Review 112 (2): 446–467. https://doi.org/10.1037/0033-
295X.112.2.446.
Kelly, Charles J. 1994. “The God of Classical Theism and the Doctrine of the
Incarnation.” International Journal for Philosophy of Religion 35 (1): 1–20.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
References 63

Kereszty, Roch A. 2002. Jesus Christ: Fundamentals of Christology. Revised


and updated 3rd Eedition. Staten Island, NY: Alba House.
Kevern, Peter. 2002. “Limping Principles: A Reply to Brian Hebblethwaite on
‘The Impossibility of Multiple Incarnations.’” Theology 105 (827):
342–347. https://doi.org/10.1177/0040571X0210500503.
King, Rolfe. 2015. “Atonement and the Completed Perfection of Human
Nature.” International Journal of Philosophy and Theology 76 (1):
69–84. https://doi.org/10.1080/21692327.2015.1043931.
Klima, Gyula. 1984. “Libellus Pro Sapiente.” New Scholasticism 58 (2):
207–219.
Labooy, G. H. 2019. “Stepped Characterisation: A Metaphysical Defence of
qua-Propositions in Christology.” International Journal for Philosophy of
Religion, January. https://doi.org/10.1007/s11153-019–09698-y.
Le Poidevin, Robin. 2009a. “Identity and the Composite Christ: An
Incarnational Dilemma.” Religious Studies 45 (2): 167–186.
2009b. “Incarnation: Metaphysical Issues.” Philosophy Compass 4 (4):
703–714.
2011. “Multiple Incarnations and Distributed Persons.” In The Metaphysics
of the Incarnation, edited by Anna Marmodoro and Jonathan Hill,
228–241. Oxford: Oxford University Press.
2013. “Kenosis, Necessity and Incarnation.” Heythrop Journal 54 (2):
214–227. https://doi.org/10.1111/j.1468–2265.2012.00796.x.
Leftow, Brian. 1989. “Necessary Moral Perfection.” Pacific Philosophical
Quarterly 70 (3): 240–260.
1991. “Timelessness and Foreknowledge.” Philosophical Studies: An
International Journal for Philosophy in the Analytic Tradition 63 (3): 309–325.
2004. “A Timeless God Incarnate.” In The Incarnation, edited by Stephen
T. Davis, Daniel Kendall, and Gerald O’Collins, 273–299. Oxford: Oxford
University Press.
2009. Time and Eternity. Ithaca, NY: Cornell University Press.
2011. “Composition and Christology.” Faith and Philosophy 28 (3):
310–322.
2015. “Against Materialist Christology.” In Christian Philosophy of
Religion: Essays in Honor of Stephen T. Davis, edited by C. P. Ruloff,
65–94. Notre Dame, IN: University of Notre Dame Press.
Lim, Joungbin. 2019. “In Defense of Physicalist Christology.” Sophia, June.
https://doi.org/10.1007/s11841-019–0718-5.
Loke, Andrew. 2009. “On the Coherence of the Incarnation: The Divine
Preconscious Model.” Neue Zeitschrift Für Systematische Theologie Und
Religionsphilosophie 51 (1): 50–63.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
64 References

2011. “Solving a Paradox against Concrete-Composite Christology:


A Modified Hylomorphic Proposal.” Religious Studies 47 (4): 493–502.
https://doi.org/10.1017/S0034412510000521.
2013. “The Incarnation and Jesus’ Apparent Limitation in Knowledge.” New
Blackfriars 94 (1053): 583–602. https://doi.org/10.1111/j.1741–2005
.2012.01500.x.
2019. The Origin of Divine Christology. Reprint edition. Cambridge:
Cambridge University Press.
Lonergan, Bernard. 2016. The Incarnate Word: Volume 8. Edited by Robert
Doran, SJ, and Jeremy Wilkins. Translated by Charles Hefling Jr. 1st
edition. Toronto: University of Toronto Press.
Luther, Martin. 1971. Luther’s Works: Word and Sacrament, 4. Edited by
Helmut T. Lehmann. Minneapolis, MN: Fortress Press.
Macquarrie, John. 1990. Jesus Christ in Modern Thought. 1st edition. London :
SCM Press.
de Margerie, Bertrand. 1980. Human Knowledge of Christ: The Knowledge,
Fore-Knowledge and Consciousness, Even in the Pre-Paschal Period, of
Christ the Redeemer. Boston: Pauline Books and Media.
Maritain, Jacques. 1969. On the Grace and Humanity of Jesus. Translated by
Joseph W. Evans. 1st edition. St. Louis, MO: Herder and Herder.
Marmodoro, Anna, and Jonathan Hill. 2008. “Modeling the Metaphysics of the
Incarnation.” Philosophy and Theology. July 1, 2008. https://doi.org/10
.5840/philtheol2008201/25.
2011. The Metaphysics of the Incarnation. Oxford: Oxford University Press.
Mascall, Eric Lionel. 1965. Christian Theology and Natural Science: Some
Questions in Their Relations. Hamden, CT: Archon Books.
Mawson, T. J. 2018. The Divine Attributes. Cambridge: Cambridge University
Press.
McCall, Thomas H. 2019. Against God and Nature: The Doctrine of Sin.
Wheaton, IL: Crossway.
McDaniel, Kris. 2017. The Fragmentation of Being. Oxford: Oxford University
Press.
McFarland, Ian. 2007. “‘Willing Is Not Choosing’: Some Anthropological
Implications of Dyothelite Christology.” International Journal of
Systematic Theology 9 (1): 3–23.
2008. “Fallen or Unfallen? Christ’s Human Nature and the Ontology of
Human Sinfulness.” International Journal of Systematic Theology 10
(4): 399–415. https://doi.org/10.1111/j.1468–2400.2008.00382.x.
McGinnis, Andrew M. 2014. The Son of God beyond the Flesh: A Historical
and Theological Study of the Extra Calvinisticum. Bloomsbury.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
References 65

McKinley, John Elton. 2009. Tempted for Us: Theological Models and the
Practical Relevance of Christ’s Impeccability and Temptation. Eugene,
OR: Paternoster.
2011. “Four Patristic Models of Jesus Christ’s Impeccability and
Temptation.” Perichoresis 9 (1): 29–66.
2015. “A Model of Jesus Christ’s Two Wills in View of Theology Proper and
Anthropology.” Southern Baptist Journal of Theology 19 (1): 69–89.
McNabb, Tyler Dalton. 2018. Religious Epistemology. Cambridge: Cambridge
University Press.
Merricks, Trenton. 2007. “The Word Made Flesh: Dualism, Physicalism, and
the Incarnation” In Persons: Human and Divine, edited by Peter van
Inwagen and Dean Zimmerman, 281–301. Oxford: Oxford University
Press.
Milyavskaya, Marina, Michael Inzlicht, Nora Hope, and Richard Koestner.
2015. “Saying ‘No’ to Temptation: Want-to Motivation Improves Self-
Regulation by Reducing Temptation Rather than by Increasing Self-
Control.” Journal of Personality and Social Psychology 109 (4):
677–693. https://doi.org/10.1037/pspp0000045.
Moloney, Raymond. 2000. Knowledge of Christ. 1st edition. London;
Bloomsbury Academic.
Moreland, J. P., and William Lane Craig. 2003. Philosophical Foundations for
a Christian Worldview. Downers Grove, IL: IVP Academic.
Morris, Thomas V. 1983. “Impeccability.” Analysis 43 (2): 106–112.
1986. “Perfection and Power.” International Journal for Philosophy of
Religion 20 (2/3): 165–168.
1987. The Logic of God Incarnate. Ithaca, NY: Cornell University Press.
2009. “The Metaphysics of God Incarnate.” In Oxford Readings in
Philosophical Theology, edited by Michael C. Rea and Thomas P. Flint,
211–224. Oxford: Oxford University Press.
Mullins, R. T. 2016. The End of the Timeless God. Oxford: Oxford University
Press.
Murray, Michael, and Michael C. Rea. 2008. An Introduction to the Philosophy
of Religion. 1st edition. Cambridge: Cambridge University Press.
Nevitt, Turner C. 2016. “Aquinas on the Death of Christ.” American Catholic
Philosophical Quarterly 90 (1): 77–99.
O’Collins, Gerald. 1995. Christology: A Biblical, Historical, and Systematic
Study of Jesus Christ. Oxford: Oxford University Press.
2002. “The Incarnation: The Critical Issues.” In The Incarnation, edited by
Stephen T. Davis, Daniel Kendall, and Gerald O’Collins, 1–30. Oxford:
Oxford University Press.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
66 References

Oord, Thomas Jay, William Hasker, and Dean Zimmerman. 2011. God in an Open
Universe: Science, Metaphysics, and Open Theism. Eugene,OR: Pickwick.
Page, Ben. 2019. “Wherein Lies the Debate? Concerning Whether God Is a
Person.” International Journal for Philosophy of Religion 85 (3):
297–317. https://doi.org/10.1007/s11153-018–9694-x.
Pawl, Timothy. 2012. “Transubstantiation, Tropes and Truthmakers.” American
Catholic Philosophical Quarterly 86 (1): 71–96.
2014a. “A Solution to the Fundamental Philosophical Problem of
Christology.” Journal of Analytic Theology 2: 61–85.
2014b. “The Freedom of Christ and Explanatory Priority.” Religious Studies
50 (2): 157–173. https://doi.org/10.1017/S0034412513000309.
2014c. “The Freedom of Christ and the Problem of Deliberation.”
International Journal for Philosophy of Religion 75 (3): 233–247.
2015. “Conciliar Christology and the Problem of Incompatible Predications.”
Scientia et Fides 3 (2): 85–106.
2016a. “Brian Hebblethwaite’s Arguments against Multiple Incarnations.”
Religious Studies 52 (1): 117–130. https://doi.org/10.1017/S00344125
14000626.
2016b. “Temporary Intrinsics and Christological Predication.” In Oxford
Studies in Philosophy of Religion, Volume 7, edited by Jonathan
L Kvanvig, 157–189. Oxford: Oxford University Press.
2016c. “Thomistic Multiple Incarnations.” Heythrop Journal 57 (2):
359–370. https://doi.org/10.1111/heyj.12230.
2016d. In Defense of Conciliar Christology: A Philosophical Essay. 1st
edition. Oxford: Oxford University Press.
2018. “Conciliar Christology and the Consistency of Divine Immutability
with a Mutable, Incarnate God.” Nova et Vetera 16 (3): 913–937.
2019a. “Explosive Theology: A Reply to Jc Beall’s ‘Christ –
A Contradiction.’” Journal of Analytic Theology 7: 440–451.
2019b. In Defense of Extended Conciliar Christology: A Philosophical
Essay. Oxford: Oxford University Press.
2020. “The Metaphysics of the Incarnation: Christ’s Human Nature.” In
Herausforderungen Und Modifikationen Des Klassischen Theismus,
edited by Thomas Marschler and Thomas Schärtl, forthcoming. Munster:
Aschendorff.
Pawl, Timothy, and Mark Spencer. 2016. “Christologically Inspired,
Empirically Motivated Hylomorphism.” Res Philosophica 91 (1):
137–160.
Pawl, Timothy, and Kevin Timpe. 2009. “Incompatibilism, Sin, and Free Will in
Heaven.” Faith and Philosophy 26 (4): 398–419.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
References 67

2013. “Heavenly Freedom: A Response to Cowan.” Faith and Philosophy 30


(2): 188–197.
2016. “Freedom and the Incarnation.” Philosophy Compass 11 (11):
743–756.
2017. “Paradise and Growing in Virtue.” In Paradise Understood, 97–109.
Oxford: Oxford University Press.
Pelser, Adam. 2019. “Temptation, Virtue, and the Character of Christ.” Faith
and Philosophy: Journal of the Society of Christian Philosophers 36 (1).
https://place.asburyseminary.edu/faithandphilosophy/vol36/iss1/5.
Plantinga, Alvin. 1999. “On Heresy, Mind, and Truth.” Faith and Philosophy 16
(2): 182–193.
Pohle, Joseph. 1911. The Divine Trinity: A Dogmatic Treatise. St. Louis, MO:
B. Herder.
1913. Christology: A Dogmatic Treatise on the Incarnation. St. Louis, MO:
B. Herder.
Rea, Michael. 2011. “Hylomorphism and the Incarnation.” In The Metaphysics
of the Incarnation, edited by Anna Marmodoro and Jonathan Hill,
134–152. Oxford: Oxford University Press.
Rea, Michael C. 2003. “Relative Identity and the Doctrine of the Trinity.”
Philosophia Christi 5 (2): 431–445. https://doi.org/10.5840/pc20035247.
Rebenich, Stefan. 2002. Jerome. London; Routledge.
Rhoda, Alan. 2007. “The Philosophical Case for Open Theism.” Philosophia 35
(3–4): 301–311.
2008. “Generic Open Theism and Some Varieties Thereof.” Religious Studies
44 (2): 225–234.
2011. “The Fivefold Openness of the Future.” In God in an Open Universe,
edited by William Hasker, Thomas Jay Oord, and Dean Zimmerman,
69–93. Eugene, OR: Pickwick.
Riches, Aaron. 2016. Ecce Homo: On the Divine Unity of Christ. Grand Rapids,
MI: Eerdmans.
Rogers, Katherin A. 2010. “Incarnation.” In The Cambridge Companion to
Christian Philosophical Theology, edited by Charles Taliaferro and Chad
V. Meister, 95–107. Cambridge: Cambridge University Press.
2013. “The Incarnation As Action Composite.” Faith and Philosophy.
August 1, 2013. https://doi.org/10.5840/faithphil201330324.
2016. “Christ’s Freedom: Anselm vs Molina.” Religious Studies FirstView
(July): 1–16. https://doi.org/10.1017/S0034412516000093.
Rosenberg, Randall S. 2010. “Christ’s Human Knowledge: A Conversation
with Lonergan and Balthasar.” Theological Studies 71 (4): 817–845.
https://doi.org/10.1177/004056391007100403.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
68 References

Rota, Michael. 2010. “The Eternity Solution to the Problem of Human Freedom
and Divine Foreknowledge.” European Journal for Philosophy of Religion
2 (1): 165–186.
Sanders, John. 1998. The God Who Risks: A Theology of Providence. Downers
Grove, IL: InterVarsity Press.
Scarpelli, Therese. 2007. “Bonaventure’s Christocentric Epistemology: Christ’s
Human Knowledge As the Epitome of Illumination in De Scientia Christi.”
Franciscan Studies 65 (1): 63–86.
Schmaus, Michael. 1971. Dogma 3: God and His Christ. 1st edition. Mission,
KS: Sheed and Ward.
Senor, Thomas D. 2002. “Incarnation, Timelessness, and Leibniz’s Law
Problems.” In God and Time: Essays on the Divine Nature, edited by
Gregory E. Ganssle and David M. Woodruff, 220–235. Oxford: Oxford
University Press.
2011. “Drawing on Many Traditions: An Ecumenical Kenotic Christology.”
In The Metaphysics of the Incarnation, edited by Anna Marmodoro and
Jonathan Hill, 88–113. Oxford: Oxford University Press.
Sharpe, Kevin W. 2017. “The Incarnation, Soul-Free: Physicalism, Kind
Membership, and the Incarnation.” Religious Studies 53 (1): 117–131.
https://doi.org/10.1017/S0034412515000530.
Sollier, Joseph. 1907. “Apollinarianism.” In The Catholic Encyclopedia.
New York: Robert Appleton. www.newadvent.org/cathen/01615b.htm.
Spence, Alan. 2008. Christology: A Guide for the Perplexed. London; T &
T Clark.
Stump, Eleonore. 1989. “Review of Morris’ The Logic of God Incarnate.” Faith
and Philosophy 6: 218–223.
1999. “Orthodoxy and Heresy.” Faith and Philosophy 16 (2): 147–163.
https://doi.org/10.5840/faithphil199916217.
2004. “Aquinas’s Metaphysics of the Incarnation.” In The Incarnation, edited
by Stephen T. Davis, Daniel Kendall, and Gerald O’Collins, 197–218.
Oxford: Oxford University Press.
2005. Aquinas. New York: Routledge.
2019. Atonement. Oxford: Oxford University Press.
Stump, Eleonore, and Norman Kretzmann. 1981. “Eternity.” Journal of
Philosophy 78 (8): 429–458. https://doi.org/10.2307/2026047.
1991. “Prophecy, Past Truth, and Eternity.” Philosophical Perspectives 5:
395–424. https://doi.org/10.2307/2214103.
Sturch, Richard. 1991. The Word and the Christ: An Essay in Analytic
Christology. Oxford; Oxford University Press.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
References 69

Swinburne, Richard. 1993. The Coherence of Theism. Revised edition. Oxford:


Oxford University Press.
1994. The Christian God. 1st edition. Oxford: Oxford University Press.
Tamburro, Richard. 2014. The Free Actions of Glorified Saints. York:
University of York.
Tanner, Norman P. 1990. Decrees of the Ecumenical Councils 2 Volume Set.
Washington, DC: Georgetown University Press.
2001. The Councils of the Church: A Short History. New York: Crossroad.
Thompson, Thomas R. 2006. “Nineteenth-Century Kenotic Christology: The
Waxing, Waning, and Weighing of a Quest for a Coherent Orthodoxy.” In
Exploring Kenotic Christology, edited by C. Stephen Evans, 74–111.
Oxford: Oxford University Press.
Tilling, Chris. 2015. Paul’s Divine Christology. Eerdmans.
Tuggy, Dale. 2007. “Three Roads to Open Theism.” Faith and Philosophy 24
(1): 28–51.
Turcescu, Lucian. 2005. Gregory of Nyssa and the Concept of Divine Persons.
New York: Oxford University Press.
Turner, James T. 2017. “On Two Reasons Christian Theologians Should Reject
the Intermediate State.” Journal of Reformed Theology 11 (1–2): 121–139.
https://doi.org/10.1163/15697312–01101023.
2019. “Hylemorphism, Rigid Designators, and the Disembodied ‘Jesus’:
A Call for Clarification.” Religious Studies, March, 1–16. https://doi.org
/10.1017/S0034412519000040.
Twombly, Charles C. 2015. Perichoresis and Personhood: God, Christ, and
Salvation in John of Damascus. Eugene, OR: Pickwick.
Vallicella, William F. 2002. “Incarnation and Identity.” Philo 5 (1):
84–93.
Van Inwagen, Peter. 1998. “Incarnation and Christology.” In Routledge
Encyclopedia of Philosophy, edited by E. Craig. London: Routledge. www
.rep.routledge.com/article/K038SECT4.
Van Horn, Luke. 2010. “Merricks’s Soulless Savior.” Faith and Philosophy:
Journal of the Society of Christian Philosophers 27 (3): 330–341.
Ward, Keith. 1998. God, Faith, and the New Millennium: Christian Belief in an
Age of Science. Oxford: Oneworld.
Ware, Bruce A. 2013. The Man Christ Jesus: Theological Questions on the
Humanity of Christ. Wheaton, IL: Crossway.
Weinandy, Thomas G. 2004. “Jesus’ Filial Vision of the Father.” Pro Ecclesia
13 (2): 189–201.
2014. Jesus: Essays in Christology. Ave Maria, FL: Catholic University of
America Press.
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
70 References

Weis, E. A. 2003. “Impeccability of Christ.” In New Catholic Encyclopedia,


2nd edition, 7:361. Detroit, MI: Gale.
Wellum, Stephen J. 2016. God the Son Incarnate: The Doctrine of Christ.
WheatonIL: Crossway.
Wesche, Kenneth Paul. 1997. On the Person of Christ: The Christology of
Emperor Justinian. Crestwood, N.Y: St Vladimirs Seminary Press.
Werther, David. 1993. “The Temptation of God Incarnate.” Religious Studies 29
(1): 47–50. https://doi.org/10.2307/20019589.
2005. “Divine Foreknowledge, Harry Frankfurt, and ‘Hyper-
Incompatibilism.’” Ars Disputandi 5 (1): 79–85. https://doi.org/10.1080
/15665399.2005.10819868.
2012. “Freedom, Temptation, and Incarnation.” In Philosophy and the
Christian Worldview: Analysis, Assessment and Development, edited by
David Werther and Mark Linville, 252–64. New York: Continuum.
Wessling, Jordan. 2013. “Christology and Conciliar Authority.” In Christology:
Ancient & Modern, edited by Oliver D. Crisp and Fred Sanders. Grand
Rapids, MI: Zondervan.
White, Thomas Joseph. 2005. “The Voluntary Action of the Earthly Christ and
the Necessity of the Beatific Vision.” The Thomist 69 (4): 497–534.
2016. The Incarnate Lord: A Thomistic Study in Christology. Washington,
DC: Catholic University of America Press.
Williams, C. J. F. 1968. “A Programme for Christology.” Religious Studies 3
(2): 513–524.
Willis, E. David. 1966. Calvin’s Catholic Christology: The Function of the So-
Called Extra Calvinisticum in Calvin’s Theology. Brill.
Zagzebski, Linda Trinkaus. 1991. The Dilemma of Freedom and
Foreknowledge. 1st edition. New York: Oxford University Press.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
Acknowledgments
I thank those who participated in a workshop on the manuscript of this book at
the Logos Institute for Analytic and Exegetical Theology (University of St
Andrews): Dennis Bray, Joshua Cockayne, Preston Hill, Matthew Joss, Carol
King, Derek King, Christa McKirland, Jonathan Rutledge, Andrew Torrance,
and Koert Verhagen. I thank the Logos Institute for a senior research fellowship
(2019), during which this book was written, and the Templeton Religious Trust
(ID: TRT0095/58801) for funding that fellowship. I thank Jc Beall, Mark
DelCogliano, David Efird – whose recent death is a tragic loss in so many
ways – W. Matthews Grant, Hud Hudson, Noah Jones, Gavin Kerr, Ryan
Mullins, Ben Page, Faith Pawl, Michael Rea, Mark Spencer, Jim Stone,
Allison Timpe (but not Kevin Timpe), and Chandler Warren for comments on
the manuscript. I thank the John Templeton Foundation for a grant (ID: 61012),
which provided some of the research time I used in the writing of this book.
I thank Fr. George Welzbacher for reading the entire manuscript and giving
careful comments in his official role as Censor Librorum for the Archdiocese of
St. Paul and Minneapolis. I thank Fr. Welzbacher for providing a Nihil Obstat to
the book, and Most Reverend Bernard A. Hebda, Archbishop of Saint Paul and
Minneapolis, for his Imprimatur.
This book is dedicated to Faith Pawl, an excellent human. (The praise I
intended to include here was brilliant, moving, and longer than word limits
allow. Trust me: it was the sort of thing you dream of reading about yourself, and
not a bit of it was undeserved.)

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
Philosophy of Religion

Yujin Nagasawa
University of Birmingham
Yujin Nagasawa is Professor of Philosophy and Co-director of the John Hick Centre for
Philosophy of Religion at the University of Birmingham. He is currently President of the
British Society for the Philosophy of Religion. He is a member of the editorial board of
Religious Studies, the International Journal for Philosophy of Religion and Philosophy
Compass.

About the Series


This Cambridge Elements series provides concise and structured introductions to all the
central topics in the philosophy of religion. It offers balanced, comprehensive coverage
of multiple perspectives in the philosophy of religion. Contributors to the series are
cutting-edge researchers who approach central issues in the philosophy of religion. Each
provides a reliable resource for academic readers and develops new ideas and argu-
ments from a unique viewpoint.

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341
Philosophy of Religion

Elements in the Series


God and Time
Natalja Deng
Continental Philosophy of Religion
Elizabeth Burns
Religious Diversity and Religious Progress
Robert McKim
Religious Fictionalism
Robin Le Poidevin
God and Morality
Anne Jeffrey
God, Soul and the Meaning of Life
Thaddeus Metz
God and Human Freedom
Leigh C. Vicens and Simon Kittle
God and Abstract Objects
Einar Duenger Bøhn
The Problem of Evil
Michael Tooley
God and Emotion
R. T. Mullins
The Incarnation
Timothy J. Pawl

Downloaded from https://www.cambridge.org/core. IP address: 112.203.39.42, on 30 Sep 2020 at 13:53:11, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781108558341

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy