This article was downloaded by: [University of Veracruzana]

On: 05 August 2015, At: 11:18

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: 5 Howick Place,
London, SW1P 1WG

Death Studies
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/udst20

New Places of Remembrance: Individual Web Memorials

in the Netherlands
a

Mirjam Klaassens & Maarten J. Bijlsma

Department of Cultural Geography, Faculty of Spatial Sciences , University of Groningen ,

Groningen , The Netherlands
Accepted author version posted online: 26 Jul 2013.Published online: 28 Aug 2013.

Click for updates

To cite this article: Mirjam Klaassens & Maarten J. Bijlsma (2014) New Places of Remembrance: Individual Web Memorials in
the Netherlands, Death Studies, 38:5, 283-293, DOI: 10.1080/07481187.2012.742474
To link to this article: http://dx.doi.org/10.1080/07481187.2012.742474

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Death Studies, 38: 283293, 2014

Copyright # Taylor & Francis Group, LLC
ISSN: 0748-1187 print=1091-7683 online
DOI: 10.1080/07481187.2012.742474

New Places of Remembrance: Individual Web

Memorials in the Netherlands
Mirjam Klaassens and Maarten J. Bijlsma

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

Department of Cultural Geography, Faculty of Spatial Sciences,

University of Groningen, Groningen, The Netherlands

The creation of places of remembrance in virtual space constitutes a new ritual to

commemorate the dead. The purpose of this study is to explore for whom individual
Web memorials are meaningful places, who are commemorated, and whether they constitute a community of support. We analyzed 181 Dutch Web memorials and conducted
content analysis of messages posted in four guestbooks. We found that parents, and in
particular mothers, create Web memorials in remembrance of their deceased children.
The memorials provide access to a community of social support, consisting primarily
of strangers and=or people who have experienced a similar loss.

The Internet has provided a relatively new place for

bereaved people to express their grief and to memorialize
their deceased loved ones. One form of memorialization
online is through posting memorials at websites known
as Web cemeteries. These memorials primarily contain
text but in some of the virtual cemeteries it is possible
to insert pictures and to add sound. Similar to the traditional cemetery, the Web cemetery provides a place
to memorialize and visit memorials to the dead (Roberts
& Vidal, 19992000). Unlike physical cemeteries, Web
memorials allow anyone to commemorate a deceased
loved one in his or her own way, whenever and from
wherever he or she chooses, and to share memories and
information about the deceased person with others (de
Vries & Rutherford, 2004; Roberts, 2004a).
Web memorials are virtual places that represent
new opportunities for postdeath rituals (de Vries &
Rutherford, 2004). This phenomenon reects changing
attitudes toward dying and mourning in contemporary

Received 6 June 2011; accepted 11 October 2012.

We thank Sjaak Moerman who conducted the analysis of the guestbook entries as part of his Bachelors thesis together with Maarten
Bijlsma on virtual places of remembrance in 2009.
Address correspondence to Mirjam Klaassens, Department of
Cultural Geography, Faculty of Spatial Sciences, University of
Groningen, P.O. Box 800, 9700 AV Groningen, The Netherlands.
E-mail: m.klaassens@rug.nl

Western societies. Traditional religious rituals have

increasingly fallen out of favor in the course of secularization and individualization resulting in a quest for new
rituals (Wouters, 2002). More than ever, people go
beyond mourning practices and spaces provided by the
traditional authorities of the church and the state (Clark
& Franzmann, 2006). The creation of a Web memorial
enables the bereaved to express their emotions freely in
public spaces (Mellor, 1993), something often discouraged in contemporary Western societies in which death
has been sequestered (i.e., removed from the public
sphere into the private world of the individual; Giddens,
1991; Mellor & Shilling, 1993). Public access to Web
memorials makes it possible for bereaved individuals to
meet and interact with people who have experienced a
similar loss and offers a sense of empathy and support
in a virtual community of grievers (Moss, 2004; Roberts,
2004a, 2004b).
Several studies by Roberts (2004a, 2004b, 2006)
indicate that Web memorialization is experienced as
benecial in the bereavement process and is thought to
constitute a meaningful ritual. Although research on
Web memorials in the Netherlands is limited, there is a
wealth of material on this topic in the United States.
However, the primary focus of this literature is on memorials in Web cemeteries (Roberts & Vidal, 19992000).
The present study concentrates on individual Web
memorials in the Netherlands, that is, memorials that

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

284

M. KLAASSENS AND M. J. BIJLSMA

are free-standing web pages instead of those that belong

to collective web pages of Web cemeteries. We studied
this type of memorial as it is considered to be the most
personal form of Web memorialization (Bijlsma &
Moerman, 2009; Roberts, 2004a). The construction of
an individual Web memorial provides mourners freedom
to express their grief as they wish, because no standard
template is used, unlike virtual cemeteries (Roberts,
2004a). Consequently, studying individual Web memorials provides a better understanding of the way grief is
expressed and manifested online.
The purpose of this article is to obtain a better understanding of those who create Web memorials in the
Netherlands, who are commemorated, and whether the
memorials provide access to a community of support.
In doing so, we examine certain details of those who create Web memorials (i.e., gender and relationship with
the deceased person) and for whom such memorials are
constructed (i.e., gender, age, and cause of death of the
deceased person). This differs from other studies that
focus on Web memorials created for a specic cause of
death, such as persons with AIDS (Blando, GravesFerrick, & Goecke, 2004), or particular relationships
between the living and the dead, for example deceased
mothers (Nager & de Vries, 2004) and deceased children
(Peelen & Altena, 2008). Previous studies that focus
on demographic characteristics of the memorialized
deceased use descriptive statistics (de Vries & Rutherford,
2004; Roberts, 2004b; Roberts & Vidal, 19992000). This
article contributes to the existing literature by applying a
binary logit model that identies those demographic
characteristics of the deceased that increase the likelihood
of a Web memorial being created.

WEB MEMORIALIZATION: A MEANINGFUL

POSTDEATH RITUAL
Several reasons are presented in the literature to explain
why Web memorials are experienced as meaningful
places of remembrance for their creators (Roberts,
2006). First, the creation and updating of online memorials involve writing about and to the deceased person,
which can be experienced as helpful and therapeutic
(Lattanzi & Hale, 19841985). That is, the act of writing
enables mourners to free themselves emotionally and
express what they are really feeling (Marshall, 2000).
In addition, Web memorials may allow the bereaved
to continue their relationship with the deceased person.
A typical feature is that guestbook entries and messages
at Web memorials are addressed to the dead and these
demonstrate continuing bonds between the bereaved
and deceased; an assurance that they will not be forgotten (Roberts, 2006). It is a way to sustain an ongoing
relationship with the dead, instead of breaking ties

and letting go (Klass, Silverman, & Nickman, 1996;

Walter, 1996).
The authors of Web memorials may experience a
sense of empathy and support in a community of grievers; such Web memorials are sites where they can share
about their loved one and their loss (Moss, 2004;
Roberts, 2004a, 2004b). The very personal stories of
the bereaved persons become public as the Internet integrates personal and mass media and enables the
bereaved (visitors and authors) to take part in two-way
communication. For example, messages of sympathy or
condolences are sent to the bereaved through the email
link provided, or by posting in the guestbook that is
often provided at Web memorials. These means of communication lead to an expansion of the support network
to include bereaved strangers who have experienced a
similar loss, and=or to the maintenance of existing
relationships. Consequently, Web memorials provide a
place for the bereaved to come together and to form a
community of support (Roberts, 2004a, 2004b, 2006).
In the face of unwritten contemporary social norms
concerning who is entitled to grieve, Web memorialization also offers an alternative ritual for a group of
grievers who are excluded from traditional grief and
mourning rituals (de Vries & Rutherford, 2004). Doka
(1989) identied these grievers as members of a disenfranchised group, who in many cases have no recognizable kin ties with the deceased, and whose loss is not
socially recognized; these mourners include friends, parents who have lost their children through miscarriage,
and pet owners. By creating a Web memorial, these disenfranchised mourners have equal access to a supportive
community (Roberts, 2006).
The duration of grief is another aspect for which
unwritten social norms are constructed. It is generally
expected that within one calendar year or less, the
bereaved will be over the loss and will function as
before (de Vries & Rutherford, 2004; Stroebe, Hansson,
Schut, & Stroebe, 2008). Although cyberspace is a recent
phenomenon, the dead who are remembered and memorialized on the Web need not be forgotten. The Web
enables relatives to (re)establish their bonds with the
dead for decades to come (Roberts, 2004b).

FACTORS INFLUENCING THE CREATION OF

A WEB MEMORIAL
A search of the literature on Web memorialization
reveals several factors that determine whether memorials
in Web cemeteries are created. First, demographic
characteristics of the deceased person, such as the gender, age, and cause of death, play a role. In Web cemeteries, more males than females are memorialized
(Blando et al., 2004), in particular young males in their

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

WEB MEMORIALS

late 40s (de Vries & Rutherford, 2004; Roberts & Vidal,
19992000). The relative youth of the deceased can be
explained by the perceived tragedy of early death
(Roberts & Vidal, 19992000). Roberts and Vidal further
showed that although the single most reported cause of
death was cancer, the majority of the causes were clearly
sudden and violent such as murder, suicides, or accidents, whereas other causes may have been sudden, such
as heart attacks and strokes.
The second factor is the demographic characteristics
of those who create Web memorials. They are predominantly female (Roberts & Vidal, 19992000). An underlying explanation is that Web memorials provide an
avenue for emotional expression, including grief, which
tends to be more characteristic of women than of men
(de Vries & Rutherford, 2004). Historically, women in
particular have been identied as grievers, who wore
mourning attire and visited the grave. Walter (1994) also
suggested that men are less likely to use rituals; the same
can also be assumed of electronic rituals (de Vries &
Rutherford, 2004). However, a study by Blando and
colleagues (2004) indicated that slightly more men than
women make AIDS memorials.
This may be due to the fact that a disproportionately
large percentage of persons who have died with AIDS
in North America have been gay men, and friends and
partners of these deceased individuals likely comprise a
larger proportion of men. (Blando et al., 2004, p. 38)

In general, the creators of Web memorials are typically

younger or from the same cohort as the deceased
(Roberts & Vidal, 19992000), which might be attributed to age-based computer skills (de Vries & Rutherford,
2004).
The third factor is the relationship between the
bereaved person and the memorialized deceased.
Although memorials are written by children, friends,
grandchildren, parents, siblings, spouses, and other
family members, the majority of memorials are by children (de Vries & Rutherford, 2004; Roberts & Vidal,
19992000).
When we consider the demographic characteristics of
the deceased person that are likely to determine the
establishment of a Web memorial, such as age, gender,
and cause of death, isolating the effect of these single
characteristics may lead to a false impression of their
signicance, as some of the demographic characteristics
are interconnected. For example, the memorialization of
certain causes of death could be explained by the prevalence of such deaths in younger age groups, or among
young males, such as trafc accidents. To overcome limitations of single characteristics, we examine the mutual
dependency of all demographic characteristics of memorialized deceased persons by using a binary logit model.

285

In the next section, the different sources of data and

methodology used in this study are presented and discussed. In the results, we rst elaborate on the design
and composition of individual web memorials to acquaint
the reader with the topic of investigation and how mourners use the Internet to create meaningful places of
remembrance. Then, we discuss for whom and by whom
the Web memorials are created, and whether the bereaved
persons create a community of support through setting
up such Web memorials. Finally, we discuss our ndings
and present the limitations and conclusions of this study.

METHODOLOGY
The individual memorial websites are often posted as
free-standing web pages and therefore are more difcult
for researchers to access as compared to virtual
cemeteries where all Web memorials can be accessed
from the starting page (Roberts, 2004a). However, in
exploring the World Wide Web, we found a web page
(http://rouwverwerking-inmemoriam.startpagina.nl) on
which 78 memorial authors had placed a link to their
free-standing websites. The individual memorial websites
themselves were very helpful in locating additional sites,
as many contained links to others, which in turn offered
links to other individual Web memorials. To enhance the
representativeness of our data set and to mitigate the
snowball effect, we collected data by means of online
search engines using different key words. The data collection resulted in a database of 171 individual memorial
websites, of which 162 were made to commemorate one
deceased person, eight sites for two deceased persons,
and one for three deceased persons. The deceased who
were commemorated by a shared individual Web memorial were bonded by family ties. In our data set, a total
of 181 deceased persons were commemorated by Web
memorials.
Viewing our data collection critically, we believe that
we have gathered a fairly representative group of individual memorial websites in the Netherlands, through
the use of different search methods. From email communication with the webmaster of the web page mentioned earlier, we learned that all memorial authors
could add the link of their own memorial to the web
page http://rouwverwerking-inmemoriam.startpagina.
nl. Consequently, no selection process occurred on the
basis of demographic background characteristics of the
deceased or the cause of death. Furthermore, we used
the search engines to nd additional sites, and consequently to obtain a more representative group of memorial authors. We continued looking for data until we
were unable to locate new Web memorials.
For all the cases in our database, data from the Web
memorials, as well as data about the authors and

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

286

M. KLAASSENS AND M. J. BIJLSMA

deceased persons was included. First we added information on the design of the memorial (inclusion of
photographs, music, videos, or poems), which shows
how mourners express their grief and create a meaningful
place of remembrance according to their taste and needs.
Second, we included demographic information of the
deceased (gender, age, and cause of death), and characteristics of those who created Web memorials (gender
and relationship to deceased). Furthermore, information
about whether a memorial provided the means of interaction with an audience (provision of a guestbook, links
to other individual Web memorials or prole accounts)
was noted in the database. This was taken as an indication of the opportunities that the memorial offered to
build a community of support.
In addition to the constructed database, we used
another source of datamessages (n 1206) posted in
guestbooksto gain more insight into those who visit
Web memorials. We selected the guestbooks of four
individual memorial sites based on different criteria.
At the beginning of our selection procedure, we selected
the guestbooks that contained a sufcient number of
entries for analysis, which was set at 100 messages.
These guestbooks made it possible to study the use of
the guestbook over a period of time. Then, we chose
the guestbooks that provided information about the
date on which the messages were posted as well as the
name of the poster (to ascertain, where possible, the gender of the visitor). Finally, the websites were selected as
the persons memorialized died have different genders
and ages and they died because of different causes: conditions originating in the perinatal period (two websites
dedicated to a newborn and stillborn baby girls), trafc
accident (one website dedicated to a 19-year-old boy),
and suicide (one website dedicated to a 16-year-old girl).
The guestbooks were analyzed for their content, characteristics of the visitors, and visitors relationship with
the author.
To obtain more insight into the contents of the guestbooks of those four Web memorials, the messages were
coded using MAXQDA, a software program designed
for text analysis. The entries were coded according to
the relationship between visitors and the bereaved. The
categories used were direct family (e.g., parents, siblings,
grandparents); acquaintances (e.g., friends, classmates,
other people known to the bereaved); people who had
experienced a similar loss; strangers (identied on the
basis of messages that stated that the visitor did not
know the author); and people whose relationship could
not be identied. The names of the visitors and the date
of posting were used to establish the frequency of visits
to the guestbook and by whom, over time. In one guestbook the gender of the guestbook posters was indicated
by each posting, in the other guestbooks it was determined where possible from their names.

The guestbook messages were analyzed using

thematic content analysis to quantify the themes under
consideration. An initial set of codes was established
based on the work of de Vries and Rutherford (2004).
Following a primary examination of the messages, other
themes were formulated. A pilot coding procedure was
applied, with three researchers coding 20 guestbook
entries independently. The codes of these messages
were compared and consequently the categories were
rened and=or extended. This resulted in the following
themes.
. Expression of sadness over the death or missing

.
.

the deceased: Recollections, expressions of loss or

emphasizing the absence of the deceased, as
well as emotions described by visitors (e.g.,
Tears rolled down my cheeks when reading your
story).
Cause of death: The cause of death is mentioned or
the circumstances in which the person died are
described.
References to God: References to God, heaven, or
religious rituals.
Reunion: References to the deceased who lives on
and watches over the bereaved. For example:
Im sure that [name of deceased] . . . proudly looks
down at us.
Expressions of support or sympathy: Declarations
of admiration for the way the bereaved deal with
the situation as well as more general expressions
of support and sympathy.
Interaction: Expressions resulting from some kind
of communication or contact; by way of thanks
for a guestbook entry or as a response to earlier
contact or messages in the guestbook.

There are several ethical considerations to keep in

mind when analyzing Web memorials. Web memorials
and guestbooks provide rich and unobtrusive data. The
accounts are very personal and idiosyncratic, and,
according to Roberts (1999), they provide a better
way of studying the impact of death on bereaved
people and their grieving process than other unobtrusive measures such as obituaries or gravestones.
The method of unobtrusive research does not require
the researcher(s) to elicit information directly from the
research subjects. Because the data is accessible to
everyone who uses the Internet and because the
accounts are placed in the public domain by the mourners themselves, we did not feel constrained in using
the data without the consent of those who created
the memorials. Nevertheless, because of the highly personal nature of the communication, we made sure that
the anonymity of the deceased persons and the authors
was preserved.

WEB MEMORIALS

RESULTS

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

The Design and Composition of Individual Web

Memorials
The primary reason for creating individual Web memorials is to remember the deceased person, captured in
phrases such as in loving memory or in memoriam.
Other reasons expressed by the authors are to share their
personal stories, experiences, and emotions with others.
This has helped them in their bereavement and they
hope that it would provide comfort and support for
others as well.
The design and composition of the Web memorials
demonstrate that such memorials are highly personalized
places of remembrance and that they provide detailed
information about the deceased person as well as the
bereaved authors. All websites contain the name of the
deceased. Other very common features are the dates of
birth and death (93.1% and 97.1%, respectively) and a
picture of the deceased when he or she was alive
(98.8%), while a few Web memorials carry pictures of
the deceased after death as well. Music accompanied
38.7% of the websites, which was automatically activated
upon visiting the site, or a particular section of the site.
The sites for children often contained instrumental childrens songs. In other cases, it was the favorite music
of the deceased or it was the music that was played at
the funeral. Fourteen percent of the websites contain
video les of the deceased when he or she was still alive.
The most commonly used background color of individual Web memorials is black or blue with white stars
or clouds, possibly representing a kind of afterlife. Often
used symbols are white stars, doves, angels, candles,
roses, and hearts. These symbols are occasionally mixed
with more personal symbolism; for example, a family
with Scottish roots displayed their clans tartan on the
site. Websites dedicated to deceased children deserve
special mention, as these websites differed in their presentation. Websites for children are more colorful. This
corresponds with the design of childrens graves at cemeteries that differs from the graves of adults in the use of
color (Hallam & Hockey, 2001). The most commonly
used symbols are butteries, teddy bears, balloons, or
characters from childrens books or television programs.
At some memorials, visitors can place a message in the
guestbook with a virtual bouquet of owers, or with a
lighted candle for the deceased person. These rituals indicate that although the virtual world provides a new place
of remembrance, the performed postdeath rituals clearly
borrow elements from traditional rituals, such as placing
owers at the grave or lighting a prayer candle which is a
common practice in the Catholic Church. The creation of
a virtual memorial does not mean that traditional places
of remembrance have been replaced, because we found

287

that at least 61% of the authors had physical places of

remembrance too, as their virtual memorial carried
pictures of a grave, urn and=or roadside memorial.
This suggests that the bereaved who created a Web
memorial were looking for additional channels to articulate their grief.
The Deceased Who Are Memorialized by
Web Memorials
To understand for whom web memorials are created, we
rst determined the age, gender, and cause of death of
the deceased honored by a Web memorial. Then, we
examined the combined effects of age, gender, and cause
of death by applying a binary logit model.
To compare the deaths memorialized on the Internet
with all deaths in the Netherlands, the age of the
deceased persons for whom a Web memorial is created
is set against the distribution of the age of those who died
in the Netherlands in the same period (see Figure 1).
Figure 1 demonstrates that the age distribution of those
honored by a Web memorial is positively skewed, as they
are set up for the young, whereas the age distribution of
the deceased Dutch population is negatively skewed,
because in general deaths occur in old age. Even though
the Web memorials were dedicated to the deceased ranging from 0 to 89 years old, the average age of death
was 20 years (SD 19.8), which is greatly inuenced by
the 04 age category. This is because the most memorials
are created for those in that age group. With regard to
the gender of the deceased persons honored by a Web
memorial, we found that the percentage of males is
slightly higher (52% for males and 48% for females).
Our results further showed differences between the
distribution of causes of death in the entire Dutch population and the distribution of the population honored by
an individual Web memorial. From Table 1 it is clear
that causes that occur at a very young age such as conditions originating in the perinatal period and congenital
anomalies are overrepresented in the individual Web
memorials, reecting the causes of death of the youngest
age group (see Figure 1). Other causes of death that are
overrepresented in Web memorials are trafc accidents,
cancer, and murder. Although diseases of the circulatory
system are the most common type of death in the Netherlands, these are underrepresented in Web memorials.
To understand the relative importance of each
demographic characteristic, we applied a binary logit
regression analysis. For the dependent variable we used
a dichotomous variable measuring whether a Web memorial was created. This variable is dened in such a way
that having a Web memorial is coded 1 and having no
Web memorial is coded 0. The cases that score 0 on the
dependent variable are all deaths in the Netherlands over
the period 19972008 minus the memorialized deceased

288

M. KLAASSENS AND M. J. BIJLSMA

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

FIGURE 1 The age distribution of deceased remembered by an individual Web memorial (left) and the age distribution of the dead in the
Netherlands (right) in 19972008; The frequency scales of the two gures are different (color gure available online).

included in our database. The deceaseds gender, age,

and several causes of death are used as independent
variables (see Table 2). We inserted seven causes of death
in our regression model. The other causes as presented
in Table 1 could not be used in the model, because
some were too small in number to compare with all
deaths resulting from this cause in the Netherlands,
and others correlated highly with the variable age, such

TABLE 1
Causes of Death of the Deceased Memorialized by a Web Memorial
and the Dutch Population (19972008)

Causes of death
Trafc accident
Neoplasm (cancer)
Murder
Conditions originating in the perinatal
period
Congenital anomalies
Suicide
Diseases of the circulatory system
Other accident
Diseases of the nervous system
Chronic lower respiratory diseases
Endocrine, nutritional and metabolic
diseases
Diseases of the musculoskeletal system
and connective tissue
All othera
N

Deceased
memorialized
by Web
memorials
%

Dutch
population
%

19.5
18.9
14.5
13.8

0.7
28.8
0.3
0.3

12.6
6.3
4.4
3.1
1.9
1.3
1.3

0.4
1.1
33.3
1.7
1.7
4.7
0.6

0.4

2.5
100.1
159

25.8
99.8
1,655,432

Note: Cause of death was classied using the major categories of

the International Classication of Disease (ICD, 9th rev.) as dened
by the Statline application of Statistics Netherlands.
a
This category is classied by the authors as all other categories of
the ICD.

as conditions originating in perinatal period, which only

occurs in the youngest age category. This did not apply
to congenital anomalies and this cause of death could
therefore be used in the model.
The results of the logistic regression show that the
younger the persons age at death, the greater the likelihood that an individual Web memorial would be created.
Furthermore, the results of the logistic regression analysis show that deceased women were more likely to have
an individual Web memorial made for them than
deceased men, which will be discussed later.
The logistic regression showed that trafc accidents,
suicides, murder, congenital anomalies, and neoplasms
were signicant and increased the likelihood of having
a Web memorial created. Overall, the results of our logistic regression show that the background characteristics,
such as age and cause of death, that signicantly affect
the likelihood of the creation of a Web memorial are
bad deaths (e.g., trafc accidents, suicides, and murder); deaths that occur suddenly, unexpectedly, and=or
violently (Seale & Van der Geest, 2004). However, one
outcome of the logistic regression model does not justify
the bad or disenfranchised death explanation, as the
model shows that cancer has a signicant effect on the
likelihood that a Web memorial would be created. In
other words, when a person dies of cancer it is more
likely that a Web memorial would be created to commemorate the deceased. Although in some cases cancer
may lead to an expected and sudden death, commonly
the progression toward death is gradual, and therefore
this cause of death contains elements of what is considered a good death, that is a death that occurs at
home, with family members present, in the absence of
violence, and to a certain extent one where there is exercise of control over events (Seale & Van der Geest, 2004).
An explanation is provided by a second regression
model, in which we took into account the interaction
effects between the causes of death and the age groups

WEB MEMORIALS

289

TABLE 2
Predictors of the Creation of Web Memorials
Model 1

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

Background characteristics
of memorialized deceased
Gender (0 female)
Age
05 years
619 years
2039 years
40 years and older (reference group)
Causes of deathb
Diseases of the circulatory system
Trafc accident
Other accident
Suicide
Murder
Congenital anomalies
Neoplasms (cancer)
All other causes of death (reference group)
Interaction effects
Cancer_age2
Cancer_age3
Trafc_age2
Trafc_age3
Circulatory_age1
Circulatory_age3
Suicide_age2
Suicide_age3
Murder_age1
Murder_age2
Murder_age3
Otheraccident_age3
Constant
N
Nagelkerke R2

2 loglikelihood
Chi-square

Model 2
Exp (B)

Exp (B)

0.425 (0.163)

0.654

.449 (0.164)

0.639

5.733 (0.324)
5.177 (0.315)
4.535 (0.278)

308.910
177.112
93.235

5.736 (0.395)
4.815 (0.625)
4.651 (0.549)

309.754
123.339
104.711

0.117
1.971
0.741
1.072
2.796
0.538
1.110

(0.450)
(0.307)
(0.499)
(0.409)
(0.304)
(0.295)
(0.295)

11.385 (0.303)
1,655,432
.297
2292.731
966.966

0.890
7.181
2.098
2.922
16.385
1.712
3.033

0.000

1.027
3.963

0.172
2.093
4.695
0.541
0.762

(0.790)
(0.485)
(1.045)
(1.066)
(0.622)
(0.299)
(0.472)

0.358
52.597
0.842
8.112
109.433
1.717
2.142

1.356 (0.760)
0.140 (0.699)

1.888 (0.830)

2.286 (0.699)
1.919 (1.285)
1.688 (1.019)
0.979 (0.504)

1.424 (1.404)

3.360 (1.191)

2.260 (1.071)

1.909 (0.784)
1.495 (1.228)

11.363 (0.374)
1,655,432
.311
2245.867
1013.830

3.882
1.150
0.151
0.102
6.812
5.408
2.663
0.241
0.035
0.104
0.148
4.458
0.000

Note: Source: Data of Statistics Netherlands and authors database (19972008).

a
Dependent variable: having a Web memorial established (yes 1, no 0).
b
Whether the memorialized person died because of the particular disease (yes 1, no 0).

p < 0.1.  p < 0.05.  p < 0.01. df 1.

(see Table 2). An interaction effect occurs when the

effect of one independent variable on the dependent
variable depends on the level of a second independent
variable. We expected that the prominence of people
who died from cancer in Web memorials stems from
the young age of the deceased, and therefore we examined
the possible interaction effects on our data. This model
does not include all the interaction effects as some of
these effects consist of very small cell counts and consequently cause quasi-complete separation in the data.
We found a signicant positive interaction effect for
cancer and the second age group (619 years), which
indicates that people in that particular age group who
died because of cancer have a signicantly higher probability of having a Web memorial created for them. The
effect of the 619 years age group dominated the

outcome of cancer in the rst model. When the

interaction effects were included in the second model,
cancer does not appear as a signicant factor in setting
up a Web memorial, which supports our bad death
explanation.
The Authors of Web Memorials
The creators of individual Web memorials were generally
family members, in particular a parent or both parents of
the deceased (42%). Family groups, brothers and sisters
of the deceased together with the parents, were often
mentioned as founders of Web memorials (27%). Other
authors were partner(s) (7%), sibling(s) (7%), children
(6%), friend(s) (3%), and grandparent(s) (2%). We could
not determine the relationship in 6% of the cases. One

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

290

M. KLAASSENS AND M. J. BIJLSMA

website was designed by the deceased himself when he

was alive, in preparation for his death. This individual
suffered from a chronic disease and was often hospitalized. He wrote texts on the website for surviving relatives. After his death, the website was maintained and
updated by a family member.
The majority of the Web memorials were created by
bereaved females, although there was a sizeable group
comprising both genders. On websites made by the
parents of a deceased child or a family group, it was
typically the mother who reported about the life and
death of her deceased child, which is consistent with
ndings by Peelen and Altena (2008).
To explain the higher probability of Web memorials
being created for women as mentioned earlier, we
examined the gender of the deceased person and the
relationship between the author and the deceased by
using chi-square tests. These tests showed that children,
although they comprise a small proportion of the total
number of authors, were more likely to memorialize
females (i.e., their mothers) than males, v2 (1) 2.270,
p < 0.1. This nding is difcult to interpret as little
research has been conducted on relationship differences
in memorialization trends (Blando et al., 2004).
We examined the time that elapsed between death and
the establishment of the memorial, by using the date of
death of the deceased person and the date of rst visible
activity on the Web memorial to examine whether the
authors could belong to a disenfranchised group of
long-term grievers (Roberts, 2004b). Web memorials
provide a place to express grief long after the socially
accepted duration of one calendar year or less (de Vries
& Rutherford, 2004). We found that the majority of
the Web memorials were constructed within the rst year
after the death (55.6%). However, a considerable group
of authors could be regarded as disenfranchised grievers
as their memorials were created more than a year after
the death of their loved one (see de Vries & Rutherford,
2004). Of the memorials that were created, 65% were set
up 1 to 2 years after the person in question died, 81%
after 4 years, and 88% after 6 years. The remaining
websites were created between 6 and 11 years later.
Virtual Communities of Support
The analysis of individual Web memorials demonstrates
that the memorials incorporated several factors that
made it possible to create a network of bereaved people
and to forge a community of support. We found that
91% of the memorials contain modes of interaction, such
as a guestbook (75%) and=or email address (62%), or in a
small number of cases a link to a social network site (1%),
which enables communication to take place between
authors and visitors. Moreover, we found that 63%
of the Web memorials were connected to other Web

memorials. Websites were more likely to be linked to

the websites of persons who had died under similar
circumstances (age and=or cause of death). This appears
especially to be the case for websites dedicated to young
children. It implies that authors of Web memorials look
for support from those who had gone through similar
experiences.
To gain more insight into the virtual community, we
identied those who posted messages in the guestbooks
and their relationship with the author(s) of the Web
memorial. We coded the guestbook messages of four
Web memorials (n 1206). The majority of the messages
were posted by persons who did not know the deceased
(43%) and those who had undergone a similar experience
(25%). A small number were posted by acquaintances of
the bereaved (11%) or by surviving relatives (2%). We
found that of all the messages, 81% were written by
women, and 7% by men. Of the messages that were
posted, 7% of the messages were posted by more than
one person, while the gender of 5% of the posters could
not be determined.
An analysis of the individual guestbooks of four Web
memorials revealed some differences in those who posted
messages in the guestbook, which might be explained by
the age, gender, and cause of death of the remembered
deceased. Although the contribution of messages written
solely by men is small in every guestbook, the guestbook
of the Web memorials dedicated to the 19-year-old boy
who died in a trafc accident contained far more messages posted by men (16%) than the other guestbooks.
Approximately 10% of messages were written solely by
men in the guestbook dedicated to the 16-year-old girl
who died by suicide, whereas the percentage of the messages in the guestbooks dedicated to the newborn and
stillborn baby written solely by men was considerably
lower (3% and 2%, respectively). Furthermore, a higher
percentage of the messages posted in the guestbooks in
remembrance of the teenage boy and girl were posted
by friends, classmates, and other people known to the
bereaved (25% and 17%, respectively, compared with
9% and 2% in the guestbooks of new and stillborn)
reecting the existing social network of the teenagers
when alive.
Thematic content analysis of the messages posted in
four guestbooks revealed that the means of communication facilitated by Web memorials are used by those
who visit the guestbooks to connect with the authors.
A considerable number of messages (34.8%) showed
indications of interaction, referring either to earlier messages posted in the same guestbook or posted on another
memorial, or referring to earlier contact via e-mail. The
most common type of message was one of (emotional)
support for the memorial author(s) and his or her family
(79.4%). References to mourning or missing the deceased
were also found (20.3%) and expressed by people who

WEB MEMORIALS

did not know the deceased person but who appreciated

the tragedy of the loss. The cause of death was mentioned
in some messages (17.3%), most notably on the website
dedicated to a person who had died by suicide. In terms
of the contents of these messages, people expressed their
incomprehension as to why the deceased had chosen suicide, or, more directly, they asked why he or she had chosen to do so. Messages referring to the afterlife, religion,
or God were uncommon (2.2%), conrming the notion
that the Netherlands is one of the more secular countries
in the world (Venbrux, Peelen, & Altena, 2009).

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

DISCUSSION
The present study concentrates on individual Web memorials, which has received little attention within the
literature on Web memorialization. We found that individual Web memorials differ from those in Web cemeteries in several aspects but share commonalities as well.
First, the design and composition of the individual
Web memorials demonstrate that they are highly personalized places of remembrance. The stories, pictures,
music, and in some cases videos provide the visitor an
impression about the kind of person the deceased was.
This nding differs from memorials in Web cemeteries,
which primarily contain text (Roberts & Vidal, 1999
2000). The construction of a memorial is a way to reconstruct the identity of a deceased loved one. Some individual Web memorials invite other mourners to provide
their own memories of the deceased, so that a shared
construction of the deceased person is created (see
Rosenblatt & Elde, 1990). It is through conversations
with others who knew the deceased that a durable
biography is constructed (Walter, 1996). For the
authors, it is a way to ensure that the deceased is not forgotten. This is in line with the continuing bonds theory
offered by Klass et al. (1996), which is about integrating
the memory of the deceased into the ongoing lives of
those bereaved, instead of breaking ties between the
bereaved and the dead and letting go.
Second, our study showed that the deceased memorialized by an individual Web memorial died a bad death.
Although there are many different perceptions of what
could be considered a good or bad death across cultures and times, there are many similarities as well. In
particular, although modern societies have apparently
been successful in controlling circumstances surrounding
death, deaths that occur unexpectedly, without warning,
or violently, before a fullled life could be achieved, or
deaths that occur away from home, or without opportunity for closure, are considered bad deaths (Seale &
Van der Geest, 2004). Some of these deaths (e.g., suicide)
could be regarded as disenfranchised deaths as well.
Whereas in Western societies suicide is believed to be a

291

taboo, and consequently the surviving relatives could

be seen as disenfranchised grievers (Doka, 1989), on
the Web this cause of death appears to be less of a taboo.
This suggests that the Web memorial provides a way for
these disenfranchised grievers to emotionally express
themselves, and the explicit mentioning of the cause of
death is a way to come in contact with others who have
experienced a similar loss. The nding that individual
Web memorials commemorate bad deaths does not
differ from the study by Roberts and Vidal (1999
2000) on Web cemeteries, which indicated that although
the single most reported cause of death was cancer, the
majority of causes were clearly sudden and violent.
Third, logistic regression showed that the younger the
persons age at death, the greater the likelihood that an
individual Web memorial would be created. In our
study, the average age of the memorialized dead was
20 years, which was greatly inuenced by the 04 age
group, and which differs from ndings in other studies
on Web memorialization that report higher ages at
death. In the Web cemeteries studied by de Vries and
Rutherford (2004) and Roberts and Vidal (19992000),
the average age was late 40s. A possible explanation
for the difference in average age of the memorialized
dead by individual Web memorials and memorials in
Web cemeteries is the relationship between the author
of the Web memorial and the deceased person, which
is discussed below.
Fourth, our ndings show that parents, and in particular mothers, create a Web memorial to remember
and to commemorate their deceased child. The death
of a child can be regarded as a disenfranchised death.
It is often stated that the loss of a small child is underestimated in contemporary society, because the parents
grief seems disproportionate to the lifespan of the
deceased (de Vries, Dalla Lana, & Falck, 1994; Doka,
1989). This nding differs from other studies that show
that most memorials in Web cemeteries were founded
by children (de Vries & Rutherford, 2004; Roberts &
Vidal, 19992000). Nevertheless, these studies suggest
that children are perhaps disenfranchised grievers as well
as the death of a parent is seen as expected, timely, fair,
and less tragic (De Vries & Rutherford, 2004, p. 19).
Doka (2008) provided a possible explanation, by
suggesting different typologies of disenfranchised grief.
The loss felt by parents who mourn the death of newborn
or stillborn babies is thought to be one whereby the loss
is not acknowledged (p. 231). Little research has been
conducted on the cause of death of parents being memorialized on Web cemeteries. These causes could contain
elements of what is considered in Western cultures as
bad deaths. Further research is required into the cause
of death, to obtain more insight into the memorialized
deceased of both types of Web memorials and the role
memorials play in bereavement.

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

292

M. KLAASSENS AND M. J. BIJLSMA

Fifth, the present study showed that deceased women

have a higher likelihood of having an individual Web
memorial created than men, whereas studies on Web
cemeteries have indicated that men are honored by memorials (de Vries & Rutherford, 2004; Roberts & Vidal,
19992000). A possible explanation is provided by Nager
and de Vries (2004), who used attachment theory
(Bowlby, 1969, 1977) to obtain a better understanding
of adult daughters who memorialized their deceased
mothers online. This theory tries to explain how individuals connect with one another and why they establish
these connections. According to Nager and de Vries
(2004, p. 44), an attachment represents an affectional
bond that has been established between two persons, a
bond that is believed to be so robust and a connection
so strong that it endures over the lifetime and beyond.
The earliest attachment bond is that between parents,
mothers in particular, and their infant child. When a
mother dies, the implication is that not only the child
loses that attachment gure, but also the longest standing relationship with another person (Nager & de Vries,
2004). One way of continuing this bond is through the
establishment of a Web memorial.
Similarly, a study by Klaassens, Groote, and Huigen
(2009) indicated that the odds of being memorialized by
a roadside memorial in the Netherlands are higher for
females, although the coefcient was not statistically signicant. These ndings are difcult to interpret as little
research has been conducted on the way the relationship
between bereaved and memorialized deceased inuences
differences in rituals and memorialization places. A
suggestion for further research is to examine whether
the relationship and background characteristics of the
deceased, such as gender, is related to places and rituals
that may help mourners to cope with their loss.
Finally, through the creation or visiting of individual
Web memorials, access is provided to a community of
support. Although the notion of a virtual community is
proposed in studies on Web cemeteries, no research has
been conducted as to whether this is applicable to individual Web memorials as well. The formation of a community around Web cemeteries is not surprising given
the collective nature of Web cemeteries, which unites
the bereaved. Our study shows that the Internet also provides other ways to access a virtual community of support, through individual Web memorials. A difference
with the Web cemeteries is that the authors of individual
Web memorials choose whether they want to be part of a
community and with whom they form this community.
Websites were more likely to be linked to the websites
of persons who had died under similar circumstances
(age and=or cause of death). This appears especially to
be the case for websites dedicated to young children. It
implies that creators of Web memorials look for support
from those who had gone through similar experiences.

The analysis of guestbooks of four different Websites

showed that the community of support differs among the
bereaved who lost teenage children and the bereaved
who lost a newborn or a stillborn baby. The website of
the rst group was more visited by friends, classmates,
and other people known to the bereaved, so that the
Web memorials provided an additional place to remember the deceased with people he or she has known. For
the second group, the website provided access to an
additional community of support.
The study has some limitations that should be considered. As we described earlier, individual Web memorials differ from memorials in Web cemeteries in several
aspects. These differences may be explained by the type
of Web memorial or by the cultural context of the
present study. Further research on individual Web memorials in other countries is suggested to provide more
insight into possible cultural differences.
Four guestbooks were selected for thematic content
analysis. To obtain the full breadth of possible guestbook messages, we selected guestbooks with at least
100 messages. Because we did not study the average
number of guestbooks entries of all memorials, it is not
known whether they are representative of individual
Web memorial guestbooks in general. If the selected
guestbooks contain above-average numbers of guestbook entries, this might affect the interpretation of the
guestbooks in general. Nevertheless, the analysis of
guestbooks in our study is more exploratory in nature
and is intended to gain greater insight into visitors, the
content of their messages and their interaction with the
authors rather than making general statements of guestbooks of individual Web memorials in the Netherlands.

CONCLUSION
The purpose of this study was to obtain a deeper understanding of individual Web memorials in the Netherlands. We found that bereaved parents in particular
created an individual Web memorial for their deceased
child. The former could be typied as disenfranchised
grievers as the loss of a small child, especially newborn
or stillborn babies, is often underestimated. The creation
of a personal Web memorial enables mourners to
(re)construct and preserve the identity of the deceased
person and to make that person a part of their own lives.
It is a way to continue the bonds between the living and
the dead, so that the dead will not be forgotten. For parents, a Web memorial is an important means to recognize
the short existence of their child, as the memorial enables
them to construct and preserve the identity of their
deceased child. Especially when the deceased is the rst
child of a couple, the Web memorial is an instrument
that helps them to construct an identity as parents.

Downloaded by [University of Veracruzana] at 11:18 05 August 2015

WEB MEMORIALS

Individual Web memorials provide access to a community of support. This community is dominated by
women, who make up the majority of Web memorials
creators as well as the majority of visitors who leave messages in the guestbooks. Web visitors are primarily strangers and=or people who have experienced a similar loss,
who express their (emotional) support to the founder(s)
and his or her family. Overall, the results of our logistic
regression show that the background characteristics age
and cause of death (e.g., trafc accidents, suicides and
murder) signicantly affect the likelihood of the creation
of a Web memorial. This suggests that virtual Web memorial provides an additional network of support for
people commemorating a bad death. For survivors
of a bad death (i.e., a death that occurs suddenly and
unexpectedly), it is not possible to prepare for and to
come to terms with such a death. Therefore, postdeath
rituals such as writingwhich in our case concerns
Web memorialscould help the bereaved persons
address unnished business with the dead, to express
their loss, and to share it with others. This does not
mean that traditional places of remembrance have been
replaced by virtual memorials, because many of the
authors have physical places of remembrance. This suggests the idea that the bereaved who have created a
Web memorial were looking for additional channels to
articulate their grief.
REFERENCES
Bijlsma, M., & Moerman, S. (2009). Herdenken in een postgeograsche ruimte [Commemorating in post-geographical space].
Rooilijn, 43(1), 2227.
Blando, J. A., Graves-Ferrick, K., & Goecke, J. (2004). Relationship
differences in AIDS memorials. Omega, 49(1), 2742.
Bowlby, J. (1969). Attachment and loss, Vol. 1: Attachment. New York:
Basic Books.
Bowlby, J. (1977). The making and breaking of affectional bonds.
London: Tavistock Publications.
Clark, J., & Franzmann, M. (2006). Authority from grief, presence
and place in the making of roadside memorials. Death Studies, 30,
579599.
de Vries, B., & Rutherford, J. (2004). Memorializing loved ones on the
World Wide Web. Omega, 49(1), 526.
de Vries, B., Dalla Lana, R., & Falck, V. T. (1994). Parental bereavement over the life course: A theoretical intersection and empirical
review. Omega, 29(1), 4769.
Doka, K. J. (1989). Disenfranchised grief. In K. J. Doka (Ed.), Disenfranchised grief: Recognizing hidden sorrow (pp. 311). Lexington,
MA: Lexington Books.
Doka, K. J. (2008). Disenfranchised grief in historical and cultural
perspective. In M. Stroebe, R. Hansson, H. Schut & W. Stroebe
(Eds.), Handbook of bereavement research and practice: Advances

293

in theory and intervention (pp. 223240). Washington: American

Psychological Association.
Giddens, A. (1991). Modernity and self-identity. Self and society in the
late modern age. Cambridge, UK: Polity.
Hallam, E., & Hockey, J. (2001). Materializing culture: Death, memory
& material culture. Oxford: Berg.
Klaassens, M., Groote, P., & Huigen, P. P. P. (2009). Roadside
memorials from a geographical perspective. Mortality, 14(2), 187201.
Klass, D., Silverman, P. R., & Nickman, S. L. (1996). Continuing
bonds, new understandings of grief. London: Taylor & Francis.
Lattanzi, M., & Hale, M. E. (19841985). Giving grief words: Writing
during bereavement. Omega, 15, 4552.
Marshall, L. (2000). Some shadows of eternity: The internet and memorials to the dead (Unpublished paper). University of Newcastle upon
Tyne, UK.
Mellor, P. (1993). Death in high modernity: The contemporary
presence and absence of death. In D. Clark (Ed.), The sociology of
death (pp. 1130). Oxford, UK: Blackwell Publishers.
Mellor, P., & Shilling, C. (1993). Modernity, self identity and the
sequestration of death. Sociology, 27, 411432.
Moss, M. (2004). Grief on the web. Omega, 49(1), 7781.
Nager, E. A., & de Vries, B. (2004). Memorializing on the World Wide
Web: Patterns of grief and attachment in adult daughters of
deceased mothers. Omega, 49(1), 4356.
Peelen, J., & Altena, M. (2008). Voor altijd een stralende ster op het
web. Digitale herinneringen aan vroeg gestorven kinderen [Forever
a shiny star on the web. Digital memories of early deceased
children]. In E. Venbrux, M. Heesels & S. Bolt (Eds.), Rituele
creativiteit
(pp. 7588). Zoetermeer: Meinema.
Roberts, P. (1999). Tangible sorrow, virtual tributes: Cemeteries in
cyberspace. In B. de Vries (Ed.), End of life issues: Interdisciplinary
and multidisciplinary perspectives (pp. 337358). New York:
Springer.
Roberts, P. (2004a). The living and the dead: Community in the virtual
cemetery. Omega, 49(1), 5776.
Roberts, P. (2004b). Here today and cyberspace tomorrow:
Memorials and bereavement support on the web. Generations,
28(2), 4146.
Roberts, P. (2006). From my space to our space: The functions of
Web memorials in bereavement. The Forum. Association for Death
Education and Counseling, 32(4), 14.
Roberts, P., & Vidal, A. (19992000). Perpetual care in cyberspace: A
portrait of memorials on the web. Omega, 40(4), 521545.
Rosenblatt, P., & Elde, C. (1990). Shared reminiscence about a
deceased parent. Family Relations, 39, 206210.
Seale, C., & Van der Geest, S. (2004). Good and bad death: Introduction. Social Science & Medicine, 58, 883885.
Stroebe, M., Hansson, R., Schut, H., & Stroebe, W. (Eds.) (2008).
Handbook of bereavement research and practice: Advances in theory
and intervention. Washington, DC: American Psychological
Association.
Venbrux, E., Peelen, J., & Altena, M. (2009). Going Dutch: Individualisation, secularization and changes in death rites. Mortality, 14(2),
97101.
Walter, T. (1994). The revival of death. London: Routledge.
Walter, T. (1996). A new model of grief: Bereavement and biography.
Mortality, 1, 725.
Wouters, C. (2002). The quest for new rituals in dying and mourning:
Changes in the we-I balance. Body & Society, 8(1), 127.