eeiau “ones Seraaremeasrts
Mourning and the Transformation
of Object Relationships
Evidence for the Persistence of
Internal Attachments
John E. Baker, PhD
Harvard Medieat School and Cambridge Hospital
Psychounalysts from Freud up fo the present have defined the
{goal of mourning asthe detachment of Bbidinal tes from the
deceased love object. The author reviewed recent cnical and
empirical Titeratare that casts doubt oa ahis assumption by
showing that a continuing internal relationship Wilk the Tost
object is found in many bereaved individuals, These data
suggest a need (o reconceptualize the changes in object rela-
‘iooships tht occur during the process of mourning, Mourning
is seen as process of iner transformation that affects both the
images of the self and of the object in the mousres’s inner
‘world, It involves not the breaking of an object tie, but the
transformation of tht attschment into a sustaining internal
presence, which operates as an ongoing component in the
individuals intomal world,
Moiuming isa topic that has been adresed by a number of psychoan-
Iysts, beginning with Sigmund Freud. In “Mooring snd Metancholia.”
Joh 1 Bakes, PRD, Deparment of Payehiy, Harvard Medica Schoo}, and Cam-
Ig Hospi Cantuidgo, Massacbuss,
‘Carespondece conening stl ut be assed o Joh Baker,
‘Pop, 68 Leonard Seo, Belmont, Masechusets O28, econ al may bo seo
(o jebaker9 Omladspring com.55 BAKER
‘Froud (1917/1957) preseated a clear and consistent theory of mourning.
He staid thatthe mourner suffers ecause of his or her intvnal attachment
to the person who hss died and thatthe goal of moumnng is to detach those
‘eelings end attachments from the lost object. The result of the moaming
process is thatthe ego becomes dee ofits former ataehments und ready
to attach toa new, living person. A review of recent papers oa mourning
(Dietrich & Shabad, 1989; Horowitz, 1990) indicated that on these points,
most psychoanalytic thought remains consistent with Freud’s original
theorizing
‘The elegance and simplicity of Fend’s formulation, and its aceuracy
in describing many aspects of the mourning process, ae undeniable,
Nevertheless, there is oth clinical and empirical evidence that calls into
question the “detachment” aspect of the theory. Bowlby (1981), on the
basis of his comprehensive review of research on attachment and 1055,
concluded that 4 continning sense of the deceased individual’ presence
after death can be found in many healthy individuals. Pincus (1974) and
Silverman and Silverman (1979) provided caso material that supports this
view: Rubin (1984, 1985) described ways in which bereaved individuals
santain ongoing intemal relationships with love objects after their death,
He stated that “the relationship to the image of the deceased following
‘mourning remains a cirenmseribed but ongoing experience involving
recollection, imaging, and association at conscious and enconscious Fev=
chs” (Rubin, 1984, p. 340). Several eseatc studies ave ab produce
supportive data (Kiss, Silverman, & Nickman, 1996; Shuchter, 1986).
Informally, many analyst wil also admit thatthe detackatent theory is not
1 complete or fully accurate description of how their patonts mourn.
‘This article reviews theoretical issues and empirical Godings that
support the noed to reconceptualize how changes eceur in internal object
relationships duriog the process of mourning. Freud’s theory of mourning,
is discussed first, as well as more recent psychoanalytic cenuibations. This
is followed by a teview of clinical and empirical Witerature that casts doubt
fom the detachment theory by showing that a continuing internal relation-
ship to the lost object is found in bereaved individuals who otherwise
appear to be adjusting well. This suggests the need to mcdify the psycho-
analytic theory of mouming by proposing that an ongoing internal rela-
tionship tothe image of the fost object isan important asgect of suecessfil
‘mourning. Mourning is therefore defined as a process of inner transfor-
tion of both self and object images. This results in both ego and
superego identifications and in the consolidation of a pzsisting internal|MOURNING AND TRANSFORMATION 37
relationship that serves as a sustining presence for the beresved
individu,
I should be noted athe outset that this artiole seers only to the
phenomena of mourning aftr the death of a love object Mourning over
ctr types of losses is also discussed in the psychoanalytic literatuze but
is nota focus ofthis ancl
Mourning and Melancholia
Freud wrote “Mourning and melancholia” a a tine whea he fad not yet
articulated his structural theory of the psyche. As he thea conceived of
‘mental fein terms of libidinal cathexes, he defined the work of mourning
as follows: “Reality testing hus shown that the loved ebject no longer
exists, ad it proceeds to denrand that all ido shall bo withdrawn from
its attachments to that object” (Freud, 1917/1957, p. 244). Throoghout,
Freud emphasized the importance of realty testing as the impetus to
‘mouming. Later in the work, he stated that “each single one of the
‘memories snd situations of expectancy .. is met by the verdict of reality
that the object no fonger exists.” Also, because the salisfactions that come
‘rom being alive outweigh the satisfactions that come from the attachment
tothe deceased object, “ihe ego... ispersusdod. ,. to sever its attachment
to the object that has been abolished” (Pread, 1917/1957, p. 255),
Froud sotod that the bide roasts withdrawing from the object
‘which may account for the infens> pain that characterizes mourning.
However, the completion of mouming involves the detachment of emo-
tional bonds from the representation af the lost objec, stch that when all
of those bonds are detached, they leave the person “fees and uninhibited
again” (Frend, 1917/1957, p. 245).
In this same work, Freud also described hove some patients identify
with the lost objec, bat he sav this as a pathological process found in
rmefanchotia. In melancholic loss, libido is withdrawn ficm the object but
sot displaced on to another object. Instead this bide
serie neselsh an deniation ofthe eo wih se baronet ject. That
‘bp shadow ofthe eject fll spon the ogo, and theft cea hencefets bo
Jedged by a special agency, a aah twee an jet, tsorsaken objet In
this way an object lose ae tormed in an ego loc. Creu, 1N7AIS),
22)
He appears to be describing identification as an abnormal process
characteristic of melancholia, but not of normal mourniag which leads to8 BAKER
“‘withdravt of the libido from this object and a displacement of it on to
anew one” (Freud, 1917/1957, p. 249)
“The Process of Identification and Its Relevance to Mouming
‘Whezeas in “Mourning and melancholia” identification yas descxibod asa
pathological outcome of loss, in “The Ego ani the li,” Freud (1923/1961)
‘hanged this view radically. Fle stated that abough Idevtfcaion is seen
in melancholia, itis also a “common and typical” reaction that ead to
‘changes in the ego and ultimately (othe formation of character. When a
person fas to give up a sexual object, “there quite often ensues an
alferation of his ego which can only be described as a setting up ofthe
object inside the-ego, as it occurs in melancholia” (Froud, 1923/1961, p.
19). Freud even suggested that identification may be “the sole condition
under which the id ean givo up its objects” (Freud, 19241961, 7.19). He
forther posited this asa central process i the formation cf the ego stating
that “the character of th ego isa precipitate of abandoned object cathxes
and that it contains the history of those object-choives" (Fead, 1923/1961,
p. 19).
"This was a radical shift in his view on identification, a shift that
cconespouled with is development of the stuctural theory ofthe psyche
and with his formulation of ego psychology. However, it seems Frew
‘writing about normal personality development here and not about bereave-
‘ment per se. What type of identifications might occur ater a death, and
‘what would be considered normal or pathological in ths area, were not
addressed by Freud in these writings.
In subsequent years, however, there have been nxmerous analysts
‘who have studied the process of identification in great detail, both as a
general process and a8 an outgrowth of toss and mourning. The literature
‘on identification isnot reviewed horein excopt as it olates to the thesis of|
this article, The interested reader ig direoted to Loowald (1962) snd
‘Farman (1974) for reviews of the concept of idemtficatcn ad its elation
to mouming, or to Meissner (1981) for a comprehensive discussion ofthe
psychoanalytic theory of internalization.
The major trend in psychoanalytic theories of loss after Frend has
‘been o study the process of idenlification in great detail but to direct less
tention tothe fate of the object representations after a death, Whe there
has been a continaing attachment to the Tost objec, it has usually been
viewed as a sign of unresolved mourning.MOURNING AND TRANSFORMATION 2
c Contributions
Other Psychounsiy
Since Keud’s seminsl work on mourning, many axa have writen
bout the long-term effects of unesolved mouming during childhood
Deatsch, 1937; Fleming & Altschul, 1963; Jacobson, 1971, chapter 7).
‘Wolfenstein (1966, 1973) presented sever cates of wasolved mourning
in children, in which the children maintained persisting finaties that eit
paroat was sil alive somewhere and would someday be etuming them.
‘On the bass ofthese eases, Wolfenstin concluded tha children aé not
able to mour the loss ofa paret unl they have sucessfully separated
fom thee parents daring the adolescent period. (See Furman, 1974, for a
positive view on chileen’s abiliy to noum)
‘Theso seminal works on unresolved mourning have st the tone for
subsequent psychoanalytic thinking. By describing the way that reunion
fantasies are often found in individusls whose mourning is unresolved,
thoy supported the broader conclusion that healthy mouming requires the
«giving up of ll active intemal fantasies aboat the deceased ove object. As
Bowlby (1980) has noted however, dese observations about patholo
wouming have been inappropriately generalized and apple to the context
of heathy mourning and grief.
‘There aro fewer analysts who have writen about normal, nonpatho-
logical mourning process's Pollock, 1978; Siggins, 1966). Horowitz.
(1990) hae developed the most clabrate famwork for stidyig the
Fntrmal changes that occur following the death ofa lovecbject. Referring
to solxepresentaions as “‘sef-schemas" and object relat onhips a5 "ole-
relationship models," Horowitz described how afte the death of a Toved
‘one, the moumer’s “schemas” about his oF het imporant relationships
‘come into conflict. Tho individual wants to hold onto the old schemas in
which the beloved is liv, bat be or se is also confronted by eaity and
its indications that the beloved is gooe and is no longex avaiable.
Ta Horowitz's (1990) model, mourning involies a “working
through” proces in which the individ’ schemas of ef and ether come
nto Hine with the demands of reality, ad the individual sabe o accept
a newr image of him- oc herself that is an accurate vw of bis oc her
present stustion. In sition, “a net, ending sole slationship model
velop in which the selfs related tothe person in the pat but notin the
present or future excep through memories (Horowitz, 1980p. 317) Like
siost other psychoanalytic authors however, this viow of mourning em-
phasizes the detachment of almest all motional ts from the representa-@ BAKER
fioa of the deceased infividual. Tt does aot describe any role for a
continuing bond to the person who died,
‘Object Relations Theory
{A Glffecent perspective on mourning ean be found in the object eltions
theory of Melanie Klein (1940/1975), who placed Iss emphasis on