Author's Accepted Manuscript

Investigating the fit and accuracy of alleged

mediumistic writing: A Case Study of Chico
Xavier’s Letters
Alexandre Caroli Rocha PhD, Denise Paraná PhD,
Elizabeth Schmitt Freire PhD, Francisco Lotufo
Neto MD, PhD, Alexander Moreira-Almeida MD,
PhD

www.elsevier.de/endend

PII: S1550-8307(14)00108-6
DOI: http://dx.doi.org/10.1016/j.explore.2014.06.002
Reference: JSCH1951

To appear in: Explore

Cite this article as: Alexandre Caroli Rocha PhD, Denise Paraná PhD, Elizabeth
Schmitt Freire PhD, Francisco Lotufo Neto MD, PhD, Alexander Moreira-Almeida
MD, PhD, Investigating the fit and accuracy of alleged mediumistic writing: A Case
Study of Chico Xavier’s Letters, Explore, http://dx.doi.org/10.1016/j.
explore.2014.06.002

This is a PDF file of an unedited manuscript that has been accepted for publication. As a
service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting galley proof
before it is published in its final citable form. Please note that during the production
process errors may be discovered which could affect the content, and all legal disclaimers
that apply to the journal pertain.
 Title page

Investigating the Fit and Accuracy of Alleged Mediumistic Writing:

A Case Study of Chico Xavier’s Letters

Authors:
Alexandre Caroli Rocha, PhD1
Denise Paraná, PhD2
Elizabeth Schmitt Freire, PhD3
Francisco Lotufo Neto, MD, PhD4
Alexander Moreira-Almeida MD, PhD5

1
NUPES – Research Center in Spirituality and Health, School of Medicine, Federal
University of Juiz de Fora (UFJF), Juiz de Fora, Brazil; PROSER, Institute of Psychiatry,
School of Medicine, University of São Paulo (USP), São Paulo, Brazil.
2
NUPES – Research Center in Spirituality and Health, School of Medicine, Federal
University of Juiz de Fora (UFJF), Juiz de Fora, Brazil; PROSER, Institute of Psychiatry,
School of Medicine, University of São Paulo (USP), São Paulo, Brazil.
3
School of Education, University of Aberdeen, Aberdeen, Scotland, UK.
4
Institute of Psychiatry, School of Medicine, University of São Paulo (USP), São Paulo,
Brazil.
5
NUPES – Research Center in Spirituality and Health, School of Medicine, Federal
University of Juiz de Fora (UFJF), Juiz de Fora, Brazil.

Corresponding Author:
Dr. Alexandre Caroli Rocha
Email: [email protected]
Tel: +5501932572842
Av. Alexandre Cazelatto, 2689, casa E 2
Paulínia (SP), 13148-911
Brazil

Grant support:
This research was supported by a grant 2010/11047-0 from FAPESP (São Paulo Research
Foundation).

Conflicts of Interest
None declared

Word Count
5.617 words
1

ABSTRACT

Context: The study of mediumship is important because if mediumistic abilities are real, they

would provide empirical support for non-reductionist theories of the mind, thus having major

implications to our understanding of the mind-brain relationship. This study investigated the

alleged mediumship of Chico Xavier, a very prolific and influential ‘medium’ in Brazil.

Objective: To investigate the accuracy of the information conveyed in Xavier’s

‘psychographed’ letters (i.e., letters allegedly authored by a deceased personality) and to

explore the possible explanations for it.

Method: After a systematic search for Xavier’s psychographed letters we selected one set of

13 letters allegedly written by a same spiritual author (JP). The letters were initially screened

for the identification of items of information that were objectively verifiable. The accuracy of

the information conveyed by these items and the estimated likelihood of the Xavier’s access

to the information via normal means were rated using Fit and Leak scales based on

documents and interviews carried out with the sister and friends of JP.

Results: We identified 99 items of verifiable information conveyed on these 13 letters; 98%

of these items were rated as ‘Clear and Precise Fit,’ and no item was rated as ‘no Fit.’ We

concluded that normal explanations for accuracy of the information (i.e., fraud, chance,

information leakage, and cold reading) were only remotely plausible. These results seem to

provide empirical support for non-reductionist theories of consciousness.

Key words: mediumship, mind-brain relationship, bereavement, consciousness, spirituality,

survival.

2

INTRODUCTION

Mediumship is generally defined as an experience in which an individual (the so-

called ‘medium’) purports to be in communication with the deceased. Mediumistic

experiences are phenomena reported throughout human history, expressed as oracles,

prophets, and shamans, and being part of the Greek, Roman, and Judeo-Christian roots of

Western society, as well as of Tibetan Buddhism and Hinduism.1,2 In recent years there has

been a heightened public interest on mediumistic phenomena, as illustrated by the high

popularity of television shows, books, and movies exploring the topic.3

The study of mediumship is important because it has significant implications for our

understanding of the nature of the mind. In the nineteenth century, studies on mediumship

phenomena were vital to the development of theories on dissociation and the subliminal

mind.4,5 Crucially, if mediumistic abilities are real, they would provide empirical support for

non-reductionist theories of the mind, thus having major implications to our understanding of

the mind-brain relationship2.

Scientific investigation of mediumship began in the late 19th century. Many scientists

and scholars who participated in these early investigations, such as William James6,

concluded that orthodox explanations (i.e., fraud, lucky chance hits, and unconscious mind

activity) could explain much but not all the empirical evidence obtained, and most of them

came to accept the existence of extra-sensorial perception and/or survival of

consciousness.2,7,8

Although research on mediumship faded through the 20th century, there has been

renewed interest during the last decade. A number of studies investigating whether

mediumship provides evidence for anomalous information reception have been recently

published in medical and psychological journals with conflicting findings.3,9,10,11 These

studies found that mediums did not provide similar levels of veridical information compared

3

to each other and even the same medium in different occasions. Therefore, it is necessary to

carry out studies with particularly gifted mediums, those who have consistently and reliably

provided evidence of anomalous information reception. This approach is in line with William

James’ suggestion that mediumship research should focus on the “good specimen of the

class”.12

CHICO XAVIER

Chico Xavier (1910-2002) was a Brazilian medium who produced a wide range of

mediumistic phenomena and is considered one of the most prolific and influential mediums

of the 20th century.13,14 Xavier was raised in a very poor and illiterate Catholic family in a

rural village in Brazil. He received only an elementary education (until fourth grade) and

started to work at eight years of age on a local weaving mill. Throughout his life, Xavier

produced, allegedly by ‘psychography’ (writing under the influence of a deceased person),

more than 450 books covering a wide range of genres and styles: novels, poetry, children’s

books, short stories, letters, and essays on scientific and philosophical topics. His first

published book, a collection of 60 mediumistic poems allegedly authored by 14 best-known

Brazilian and Portuguese deceased poets, was published in 1932, and became a literary

sensation in Brazil.15

A number of archived film footages show Xavier at work: He would pick up a pencil,

with his right hand on a sheet of paper and his left hand over his eyes, and he would sit in

silence for a couple of minutes. Then his hand would slide over the paper at great speed,

never going over the edge of the paper. When a page was filled, a person sitting next to him

would remove the sheet and put another one in its place. In this way, Xavier would cover

about twenty large sheets of paper in few minutes.15

4

 By 2010, Xavier’s books had sold 50 million copies16 and were translated to more

than ten languages13,17. Yet despite his enormous popularity and record-breaking sales,

Xavier lived a very modest life with the small salary of his civil servant job. He never sought

nor received payment for anything he wrote as he donated all his copyrights to charities.13,17

Another important aspect of Xavier’s mediumistic work was the production of

personal messages allegedly written by deceased personalities to relatives and friends left

behind. These messages were produced in weekly public sessions where hundreds of

bereaved would attend with the hope of receiving a letter from their deceased loved ones. It

is estimated that Xavier produced 10,000 of these personal letters.14,18 Often these letters

would contain personal information about the deceased and their family, proper names,

surnames and nicknames, and detailed descriptions of their death. There are also reports,

requiring further investigation, of letters allegedly written entirely or partially in a foreign

language and with signatures similar to those of the deceased when they were alive.19,20

Chico Xavier had a major impact on Brazilian culture. In 1981, he received a

nomination for the Nobel Peace Prize14, and in 2010 the Brazilian Post Office issued a special

stamp to mark the centenary of his birth21. One of his books, Nosso Lar, sold more than two

million copies and was made into a movie in 2010 that attracted more than four million

spectators. Also in 2010, a biographical movie about Chico Xavier was watched by more than

three million people in Brazil.22

Despite the impact of Xavier’s mediumistic work, it has been subject to little research.

Rocha15 analyzed the stylistic, formal, and interpretative aspects of dozens of Xavier’s poems

allegedly written by renowned Brazilian and Portuguese authors. The analysis suggested that

the poems are not a product of simple literary imitation. Furthermore, the production of these

poems would have required a sophisticated and scholarly knowledge of these writers.

Rocha23 investigated the mediumistic writings from 12 of Xavier’s books attributed to

5

Brazilian writer Humberto de Campos (1886-1934). Rocha found an intricate and

sophisticated intertextuality in these books that could be achieved only by someone with a

scholarly knowledge of Campos’ work. Recently, Silva24 carried out a semiotic analysis of

ten letters psychographed by Xavier and found a consistent overlap and alternation of the

expression of the identities (ethos) of the deceased personalities and the medium in the

letters.

Given the limited research on Xavier’s mediumistic letters and the importance of his

work to the study of the mind-brain relationship, we carried out a study that aimed to

investigate the accuracy of the information conveyed in Xavier’s psychographed letters and

to explore the possible explanations for it.

METHODS

Case Selection

We carried out a systematic search for Xavier’s published and unpublished

psychographed letters in order to select one set of letters for this study. The selection of the

set of letters was based on the following criteria: (1) all letters are allegedly written by the

same spiritual author; (2) the letters provide a substantial amount of detailed, verifiable, and

specific information; and (3) the relatives to whom the letters were addressed are still alive

and would agree to be interviewed by the research team. Based on these criteria, for this

study we selected the set of 13 letters allegedly written by the spirit of a deceased young man,

Jair Presente (JP), who died by drowning on February 3, 1974 at the age of 24.

Material

We analyzed 13 letters, which were written between March 15, 1974 and January 13,

1979. Eight of these letters were published in books.25,26 We analyzed the original letters

which have been kept by the family of JP. In addition to these letters, other related material

6

were collected and analyzed: personal documents and writings of JP, two newspaper articles

about his death, and his biography written by his sister.

Participants

The principal participant in this study was SP, the older and only sibling of JP. His

father died in 2006 and his mother declined participation due to her old age and ill health.

Other participants were three close friends of JP and one friend of his parents.

Procedure

The 13 letters were screened for the identification of the items of information that

were objectively verifiable. After the identification of these items, two members of the

research team (ACR and DP) carried out interviews with the participants in order to inquire

about: (a) the accuracy of the information conveyed by these items and (b) all possible

ordinary means by which Xavier could have accessed this information. The interviews were

semi-structured and based on Stevenson’s guidelines27 for research on survival of

consciousness. We carried out five interviews with SP, each lasting two hours on average.

The interviews were recorded. After the interviews, the items of the letters were rated

independently by three members of the research team (ACR, DP and ESF) using the Fit and

Leak Scales (see below).

Instruments

- Fit Scale: This scale, which is an adaptation of the Arizona Mediumship-Process Scoring

System (AMPSS)3, was designed for the particular context of this study. It was devised as a

3-point scale for the ratings of the accuracy of the information provided by items in the letter.

The rating points of the Fit scale are: ‘No fit’ (0); ‘Fit dependent on interpretation or

imprecise fit’ (1); ‘Clear and precise fit’ (2).

7

- Leak Scale: The instrument was devised for the ratings of the estimated likelihood of the

medium’s access to the information via normal means (i.e., information ‘leakage’). The Leak

scale has five rating points:

Very unlikely leak (0) – very specific information that was allegedly unknown by

relatives/friends of the deceased person who had contact with the medium, or

information that was allegedly a secret held by a friend/relative and never revealed;

AND there was minimal contact of relative/friends with the medium (or their

assistants).

Rather unlikely leak (1) – information very specific or very private; AND

relative/friend is definite that information was never communicated to the medium (nor

to their assistants) AND there was minimal contact of relative/friends with medium (or

their assistants).

Unlikely leak (2) – Information very specific or very private; AND relative/friend is

definite that information was never communicated to the medium (nor to their

assistants), BUT there was more than minimal contact of relative/friends with medium

or their assistants.

Likely leak (3) – information very general, common or predictable.

Definite leak (4) - relative/friend confirms that information has been previously

communicated to the medium or to their assistants.

Analysis

The final ratings on the Fit and Leak were obtained through discussion and consensus

between the raters. In addition to the statistical analysis of the ratings on the Fit and Leak

scales we also performed a narrative analysis28 of the content of the letters and of the

additional material collected for the study (i.e., interviews, personal documents of JP,

newspapers articles, and the biography written by his sister).

8

Ethics

This research protocol received ethical approval by the University of São Paulo Ethics

Committee. SP gave written consent for her identity to be disclosed in the study publications.

RESULTS

JP and the Circumstances of His Death

JP was born on November 10, 1949, in Campinas, Brazil. His father was a tradesman

and his mother, a housewife. He had only one sibling, SP, born in 1947. JP was working on

his fourth year of mechanical engineering at one of the top universities in Brazil, Unicamp,

when he died, at the age of 24. He would also work as a school teacher and private tutor.

On Sunday, February 3, 1974, JP went on an outing to a reservoir in a place called

‘Praia Azul’ with five young friends. They had spent the previous night together at a fishing

farm nearby. Around 11am, JP dove into the water, but did not come up to the surface. His

friends called the fire fighters based at the reservoir and after a few minutes they found JP

and retrieved him from the water. The fire fighters tried to resuscitate him by performing

cardiopulmonary resuscitation and mouth-to-mouth ventilation, but unsuccessfully. JP was

pronounced dead at the scene. The friends reported that there was no use of drugs and very

little consumption of alcohol during the morning of the fatal accident.

The First Meeting with Xavier

JP’s family was devastated. Thirty days after his tragic death, one of the customers of

JP’s father gave him a book written by Xavier that he thought would help him to cope with

his grief. After reading the book, JP’s father decided to go to the city of Uberaba, 400 km

away from his city, to try to meet Xavier. Forty days after JP’s death, his family arrived at the

spiritist center in Uberaba where Xavier worked.

9

 A few hundred people would attend Xavier’s public mediumistic meetings at his

spiritist center. The attendance to these meetings was free. Every Friday afternoon, people

would gather in a queue for the opportunity to talk for a few minutes with Xavier19. Between

2pm and 6pm, Xavier would exchange a few words with each of the attendees in the queue.

After 6pm, Xavier would go to a small room at the back of the spiritist center with two

assistants. He would stay there until 12am writing homeopathic prescriptions along with brief

spiritual advice, which were then handed to the attendees. These prescriptions and messages

were allegedly authored by the spirits of physicians who communicated via Xavier’s

psychography. After Xavier finished the ‘psychography’ of these medical prescriptions, he

would go to the main hall where the attendees would have been waiting, accommodated in

chairs. Xavier would sit at a large table in front of the audience and he would produce

‘psychographed’ letters uninterruptedly for about three hours. When Xavier finished the

letters, around 3am, he would read them aloud for the audience. He would perform

psychography an average of six letters per night, each letter allegedly written by a different

deceased person19.

SP reported that when she and her parents arrived at Xavier’s spiritist center, on a

Friday afternoon, they had only a very brief exchange with him. According to her account,

after waiting in the queue, the only thing she said to Xavier was that she had lost her brother

and that her parents were devastated and would like to receive some news from him or even a

letter. SP was definite that they did not say any names or give any other information to

Xavier. SP and her parents then sat in the main hall waiting alongside the rest of the crowd

until the early hours of Saturday, when Xavier started the psychography of the letters. On that

night, Xavier produced seven letters. According to SP, after reading out the first letter, Xavier

asked the audience: “Who are the relatives of Jair Presente?” Astonished, SP and her parents

went to the front of the hall and stood up next to the medium while he read out the letter

10

allegedly written by JP. When Xavier finished the reading, he handed the manuscript to JP’s

mother.

Analysis of the First Letter

This first letter, containing 569 words, was written on 32 sheets of letter-sized bond

paper. The size of the letters was fairly big, making an average of 18 words per sheet.

According to SP, Xavier wrote this letter in a continuous and fast flow, without any pauses.

The letter, addressed to SP and JP’s parentsi, contained 17 items of verifiable information

(See Table 1). Among these items, there were three first names (‘Sueli’, ‘Jair’, and ‘Elvira’),

one surname (‘grandpa Basso’), and one date (‘Sunday’ – reference to the day of JP’s death).

Other information conveyed by the letter included a detailed description of the circumstances

of JP’s death (eight items), one reference to a postmortem family event (one item), and

references to past activities of the deceased (two items).

A qualitative analysis of the content of the letter indicates that the author sought

mainly to bring consolation and comfort to the family, begging them to keep calm and

positive, and to hold their faith in God’s unfaltering love. According to SP, the family was

convinced that JP was the author of the letter because of the level of accurate and precise

information conveyed by it, which had not been communicated to the medium.

The mean score on the Fit scale for this letter was 2 (SD = 0) since all information

conveyed in the letter was confirmed by the participants to be correct and accurate. In relation

to the Leak scale, the mean score for this letter was 0.94 (SD = 0.24). All but one item

(93.8%) were rated as ‘rather unlikely leak’ (1). One item was rated as ‘very unlikely leak’

(0) because it was information unknown to JP’s family present at the mediumistic session.

This item was a reference to a deceased personality named ‘Elvira’ allegedly present at the

side of JP while Xavier was writing the letter. JP’s mother had an aunt and godmother named

i
The letter begins with the following greeting: “My father, my mother, my dear Sueli…” [Meu pai, minha mãe,
minha querida Sueli].
11

‘Elvira,’ which is an uncommon name in Brazilii, but at the time of the session she did not

know that Elvira had died three years earlier. In a subsequent letter, the alleged author

confirms that Elvira was indeed the aunt and godmother of his mother.

The Subsequent Letters

Convinced that Xavier’s mediumship was authentic and that JP was the author of this

first letter, the family returned many times to Uberaba to attend other psychography sessions

of Xavier at the Spiritist Christian Communion. Between the years of 1974 and 1979, Xavier

psychographed 12 more letters allegedly authored by the spirit of JP (See Table 2). The

author described his condition and his activities in the spiritual world, his plans for the future,

and the difficulties involved in the mediumship communication. This set of 12 letters was

characterized by the intensive use of slang, humor, pun, and colloquial expressions. We

found this similar type of language in JP’s personal writings and SP confirmed that some of

the colloquial expressions used in the letters were typical of JP‘s vocabulary. According to

SP, these letters brought about a significant change in the family, as they helped them to cope

more positively with their grief and loss.

These 12 letters conveyed a total of 83 items of verifiable information (see Table 3).

Of these items, 13 (15.7%) conveyed information that was confirmed by the participants to be

previously known to the medium and thus rated ‘4’ on the Leak scale. We were not able to

determine the Leak score for 15 items (18.1%) because they bore information about

individuals and facts that were not related to JP and the family. Since we interviewed only

relatives and friends of JP, we were unable to assess the likelihood of information leak in

relation to these items. Forty-five items (54.2%) were rated as ‘unlikely leak’ because

although the participants were convinced that none of this information had been


ii
To get an approximate idea about the frequency of some names that appear in the letters, we consulted record
books of births (1971-1976) and deaths (1955-1979) in a registry office in the city of Campinas. From about
3,200 names consulted, there was no person named Elvira.
12

communicated to Xavier, more than minimal contact had developed between JP’s family and

Xavier after he psychographed his first letter. However, eight items (9.6%) were rated as

‘very unlikely leak’ because the information conveyed on these items was quite specific and

unknown to JP’s family. These items correspond to cases of ‘drop-in’ communication29 and

they will be discussed below. The overall mean score for these 12 letters on the Leak scale

was 2.18 (SD=1.1).

Analysis of the Fit of the 13 Letters

There were a total of 99 items of verifiable information conveyed by the 13 letters. Of

these items, 13 (15.7%) were definitely confirmed by participants to have been communicated

to Xavier via normal means, and 15 (18.1%) had undetermined Leak scores (as explained

above). Therefore, we excluded these 28 items from the posterior analysis. The mean score

on the Fit scale for the 71 remaining items was 1.97 (SD = 0.17). There was no item rated as

‘No Fit.’ Sixty-nine items (97.2% of these items) were rated as ‘Clear and Precise Fit’ and

two items (2.8%) were rated as ‘Fit dependent on Interpretation/imprecise’ (See Table 4).

These two items were cases of ‘drop-in’ communication (see below) whose identity could not

be definitely confirmed.

Most of the information conveyed by these 71 items were proper names (42%). Of the

24 names of people cited on these 71 items, three names were composed by first name and

surname (See Figure 1). All names were spelled correctly. Fourteen items (19.7%) were

descriptions of specific events with all but one being postmortem events. One example of a

postmortem event was the reference to the death of ‘Suzeley’ in a swimming pooliii. Twelve

items (16.9%) were descriptions of feelings/thoughts of family and friends and 10 items

(14.1%) were descriptions of the circumstances of JP’s death (See Table 5).

Drop-In Communications


iii
Suzeley was an acquaintance of SP. ‘Suzeley’ is also a very uncommon name in Brazil.
13

 In these 13 letters, we found three cases of ‘drop-in’ communications, that is,

situations where allegedly a deceased personality communicates via the medium without the

request of relatives or friends. In the case of JP’s letters, the author communicated that some

deceased personalities were asking him to deliver messages of comfort and solace to their

parents and relatives, although these were not present at the mediumistic sessions when the

letters were psychographed. These three cases vary in the degree of specificity of the

information transmitted and the extent to which the information was verifiable (see Table 6).

For example, in one of the cases, the letter conveyed the date of death (day, month, and year),

name and double surname of the deceased (Irineu Leite da Silva), and first names of their

parents. This information was later verified by SP who found a small death notice for Irineu

published in a local paper of her city, 400 km away from Xavier’s town, where the public

library did not receive the papers.

DISCUSSION

The information conveyed by the 13 letters allegedly authored by the deceased

personality of JP showed a high level of accuracy and fit. Excluding from the analysis the 13

items of information that have been definitely confirmed by participants to have been

communicated to Xavier via normal means, and the 15 items with an undetermined Leak

score, we found that 97.2% of the remaining items demonstrated a clear and precise fit.

Moreover, we found no item with a ‘wrong’ fit. The only two items that were not considered

to have clear fit were first names of deceased personalities unrelated to any of the persons

attending the mediumistic sessions and whose identities could not be confirmed without

doubt.

Not only was the information conveyed by this series of letters highly accurate, but it

was also very specific. As shown in Table 5, the majority of information expressed in these

14

letters consisted of proper names and surnames, kinship, objective description of events, and

precise dates. This information was so precise and specific that they could not possibly be

obtained by simple inference from previously confirmed items. Only 19.7% of the

information revealed some type of subjective experience, such as feelings, thoughts, and

personal interests. Kelly and Arcangel pointed out that a normal explanation for the apparent

fit of a medium’s statements is that they are so general or vague that “they can apply to many

people or be interpreted in a variety of ways”.10 Since the information conveyed by the

Xavier letters was very specific, it refutes the hypothesis that the fit of the information was

the result of chance alone or over-interpretation on the part of the participants biased by grief

or wishful thinking.

A weakness of this study is that the assessment of the likelihood of information

leakage was based on participants’ accounts of what happened in the mediumistic sessions

and their recollection of what information they had communicated to the medium. Since these

mediumistic sessions occurred more than 30 years ago, the participants’ recollections of the

extent of their communication with the medium might have been inaccurate or distorted.

Research on memory indicates that a person’s recollection of an event can be altered by post-

event information or by group influence, leading to long-lasting errors and even to the

creation of completely false memories.30,31 Thus, we cannot rule out the hypothesis that the

participants’ wishes that these letters were genuinely authored by their beloved JP might have

misled them to distort their memories of what they have actually communicated to the

medium before he psychographed the letters. However, in relation to the first letter

psychographed by Xavier, even if the participants’ recollection of the information

communicated to the medium was inaccurate or unreliable, it seems implausible that they

would have had the time and opportunity to communicate such a significant amount of

specific information, including the surname of JP’s grandfather (‘grandpa Basso’), as well as

15

the date and the detailed circumstances of his death. Xavier’s mediumistic sessions were

attended by a few hundred people and he would have only very brief contact with the

attendees who would be waiting in a long queue for the opportunity to exchange just a few

words with him.

Moreover, the rating points of the Leak scale were designed to compensate for this

unreliability of the participant’s recollection. The scale is asymmetrical so that the lowest

point in the scale corresponds to ‘very unlikely leak’ instead of ‘definitely no leak.’ Also,

when participants have had more than a minimal contact with the medium (in this study that

was the case after the first letter), the corresponding rating point in the Leak scale is ‘unlikely

leak’ even if the participant is adamant that no information was communicated to the

medium. Therefore, the Leak scale is a very ‘conservative’ measure that takes into account

the potential bias involved on the participant’s recollection of their contact with the medium.

However, despite the asymmetry and conservative bias of the Leak scale, when we analyzed

the 12 letters produced by Xavier after his first encounter with the family (i.e., letters two to

13), we found that 45 items of information were unlikely to have been obtained by normal

means and eight items were very unlikely to have been obtained by information leakage.

These eight items correspond to the cases of ‘drop-in’ communications, since the information

conveyed by them was unknown to everyone attending the mediumistic session.

Normal explanations for ‘mediumistic’ communications are fraud and cryptomnesia

(i.e., latent subconscious memory).8 Xavier could have read articles in the two local

newspapers of Campinas that reported JP’s death and either consciously (fraud) or

unconsciously (cryptomnesia) communicated this information in the letters. However, these

local papers did not publish any photography of JP’s family, so Xavier would have not been

able to identify JP’s parents and sister presented at the spiritst center as being his relatives.

Moreover, the possibility that Xavier would have had access to a local paper from Campinas,

16

a city 400 km away from where he lived (Uberaba), seemed to be rather remote. In addition,

the letter contains several pieces of information that were not reported by the newspapers, for

instance, the name of JP’s grandfather (“grandpa Basso”) and the fact that he was deceased.

Alternative sources for information leakage could be other family members or friends of JP.

However, we established that JP’s relatives had no family or friends in Uberaba and that they

traveled there exclusively to attend Xavier’s psychography sessions.

In relation to the drop-in communications, as discussed above, it seems very unlikely

that Xavier would have had access to the Campinas local newspaper where Irineu’s death

notice was published. Moreover, we could not confirm whether there were any death notices

published for the cases of Aníbal and Joãozinho Alves. However, the case of Aníbal could be

explained as a lucky chance hit since he was not identified by a surname. Yet the first name

‘Aníbal’ is uncommon in Braziliv and the likelihood that any Aníbal would have committed

suicide around the time the letter was written would seem rather small.

Another common orthodox explanation for mediumship phenomena is that the

medium gains information about their clients through the process of ‘cold reading.’ This is a

set of techniques in which the medium fishes for information, whether deliberately or

inadvertently, by taking visual and auditory feedback or clues from their client’s responses or

appearance in order to fabricate ‘accurate’ information.32 However, the process of cold

reading could not have been applied to the psychography of Xavier‘s letters since they were

written in a trance-state, in an uninterrupted flow and without any interaction with the

receivers of the letter.

We considered that the strength of this study lies with the large amount of information

conveyed by the letters. The fact that among 99 items of specific information there was not a

single one wrong seems remarkable. Chance alone could have explained the fit of a few

iv
According to our research at the registry office of Campinas, referred to in the preceding footnote, among
3,200 queried names, we found two people with the name Aníbal, both born over 110 years ago.
17

isolated cases of information, but it does not seem plausible that the accuracy of such a large

set of data could be explained by merely lucky chance hits. However, a limitation of this

study is that it analyzed only a very small portion of Xavier’s psychographed letters. We do

not know to what extent these 13 letters are a representative sample of the thousands of letters

produced by him or whether they are just exceptional exemplars.

Xavier psychographed these letters through very long hours of strenuous work and

without any financial gain. Although it is not possible to estimate the amount of verifiable

information conveyed by these thousands of letters, it would seem very unlikely that Xavier

would have employed fraudulent schemes (such as secretly eaves-dropping conversation) to

obtain this information as the production of these letters by fraudulent means would have

required a very sophisticated, complex and expensive structure that seems unreasonable given

the context and circumstances of Xavier’s work. He was a very poor man who never profited

from his psychographed work and always presented himself with extreme humility and

simplicity.13,14 JP’s family did not profit from the publication of these letters either, since the

copyrights were donated to charity. Therefore, it would seem unreasonable that JP’s family

would fabricate such a fraud as they would not have any objective gain from it.

If we can possibly rule out chance, fraud, information leakage, and cold reading as

plausible explanations for the accuracy of the information conveyed by Xavier’s letters, we

would have to consider the non-conventional hypothesis of extra-sensory or ‘anomalous’

information reception, including the possibility of survival of consciousness.2,33 Modern

scientific thought often embraces the materialistic view that mind or consciousness is solely

an epiphenomenon of brain processes. However, a number of contemporary philosophers of

mind have suggested34,35,36 that reductionist materialistic approaches to consciousness are

inadequate to fully explain the correlation between brain processes and mind states or why

brain processes should give rise to conscious experience at all. In addition, accumulated

18

evidence on other human experiences such as near-death experiences37 also seems to suggest

that there is an aspect of consciousness that transcends the brain.

CONCLUSION

The results of our investigation suggest that Xavier’s letters conveyed accurate and

precise information and that normal explanations for it (i.e., fraud, chance, information

leakage, and cold reading) are just remotely plausible. This study seems to yield empirical

support for non-reductionist theories of mind. Certainly more research on mediumistic

phenomenon is called for, and we recommend that further research on other psychographed

letters by Xavier be carried out in order to expand the scope of this investigation. Other

controlled studies on mediums as exceptionally gifted as Xavier are also imperative. As

Chalmers pointed out:

The possibility of explaining consciousness non-reductively remains open. This

would be a very different sort of explanation, requiring some radical changes in the

way we think about the structure of the world. But if we make these changes, the

beginnings of a theory of consciousness may become visible in the distance.34



Acknowledgments

We are grateful to Dr Silvio Seno Chibeni, Dr Homero Vallada, Guy Lyon Playfair,

Dr Etzel Cardeña, and Dr Welligton Zangari for their helpful comments on earlier versions of

this manuscript and for their encouraging support to our research.

REFERENCES

1. Cardeña E, Winkelman M. Altering consciousness: multidisciplinary perspectives. In:

History, Culture, and the Humanities. Santa Barbara, CA: Praeger; 2011: volume I.

19

 2. Moreira-Almeida A. Research on mediumship and the mind-brain relationship. In:
Moreira-Almeida A, Santos FS (Eds). Exploring Frontiers of the Mind-Brain
Relationship. New York, NY: Springer;2012:191-213.
3. Beischel J, Schwartz GE. Anomalous information reception by research mediums
demonstrated using a novel triple-blind protocol. Explore: The Journal of Science and
Healing, 2007;3:23-27.
4. Alvarado CS. Dissociation in Britain during the late nineteenth Century: the Society
for Psychical Research, 1882-1900. J Trauma & Dissociation 2002;3(2):9-33.
5. Crabtree A. From Mesmer to Freud: Magnetic Sleep and the Roots of Psychological
Healing. New Haven, CT: Yale University Press; 1993.
6. James W. Essays in Psychical Research: The Works of William James. Cambridge,
MA: Harvard University Press; 1986.
7. Braude SE. Immortal Remains: The Evidence for Life after Death. Lanham, MD:
Rowman & Littlefield; 2003.
8. Gauld A. Mediumship and Survival: A Century of Investigations. London, England:
Granada; 1982.
9. Jensen CG, Cardeña E. A controlled long-distance test of a professional medium. Eur
J Parapsychol 2009;24:53-67.
10. Kelly EW, Arcangel D. An investigation of mediums who claim to give information
about deceased persons. J Nerv Ment Dis 2011;199:11-17.
11. O’Keeffe C, Wiseman R. Testing alleged mediumship: Methods and results. Brit J
Psychol 2005;96:165-179.
12. James W. Report of the committee on mediumistic phenomena. In: G Murphy, RO
Ballou (Eds). William James on psychical research. New York: Viking Press;
1886/1960: 95-100.
13. Maior MS. As Vidas de Chico Xavier (2nd ed). São Paulo, SP: Planeta; 2003.
14. Stoll SJ. Espiritismo à Brasileira. São Paulo, SP: Edusp; Ed. Orion; 2003.
15. Rocha AC. A poesia transcendente de parnaso de além-túmulo. (Master’s thesis).
Retrieved in 2001 from Biblioteca Digital da Unicamp (code vtls000236698).
16. Langellier JP. Un homme insignifiant. Le Monde, 12/05/2010.
17. Ponsardin M. Chico Xavier: L'homme et le médium. Miami, FL: Edicei of America;
2011.

20

 18. Lewgoy B. O Grande Mediador: Chico Xavier e a Cultura Brasileira. Bauru: Edusc;
2004.
19. Severino PR. Life’s Triumph. São Paulo, SP: FE Editora; 1994.
20. Perandréa CA. A Psicografia à Luz da Grafoscopia. São Paulo, SP: Editora
Jornalística Fé; 1991.
21. Playfair GL. Chico Xavier: Medium of the Century. London: Roundtable; 2010.
22. Costa U. Os números do cinema nacional em 2010. Retrieved on January 21, 2011 from
http://jornalnh.com.br/blogs/setima-das-artes/302196/os-numeros-do-cinema-nacional-
em-2010.html Accessed 10.10.13
23. Rocha AC. O caso Humberto de Campos: Autoria literária e mediunidade. (Ph.D
thesis). Retrieved in 2008 from Biblioteca Digital da Unicamp (code vtls000443434)
24. Silva CA. As cartas de Chico Xavier: Uma análise semiótica. São Paulo, SP: Cultura
Acadêmica; 2012.
25. Xavier FC. Jovens no Além. São Bernardo do Campo, SP: GEEM; 1975.
26. Xavier FC. Somos Seis. São Bernardo do Campo, SP: GEEM; 1976.
27. Stevenson I. The investigation of reincarnation cases: suggestions for the guidance of
investigators. Charlottesville: University of Virginia School of Medicine; 2003.
28. Riessman CK. Narrative analysis. London: Sage; 1993.
29. Stevenson I. A communicator unknown to medium and sitters. J Am Soc Psychical R
1970;64:53-65.
30. Loftus EF, Pickrell JE. The formation of false memories. Psychiat Ann 1995;25:720-
725.
31. Edelson M, Sharot T, Dolan RJ, Dudai Y. Following the crowd: Brain substrates of long-
term memory conformity. Science 2011;333:108-111.
32. Rowland I. The Full Facts Book of Cold Reading. London: Ian Roland; 1998.
33. Almeder R. Death and Personal Survival: The Evidence for Life After Death.
Lanham: Rowman & Littlefield; 1992.
34. Chalmers DJ. The Conscious Mind: In Search of a Fundamental Theory. Oxford: Oxford
University Press; 1996.
35. Nagel T. Mind and Cosmos: Why the Materialist Neo-Darwinian Conception of Nature
is Almost Certainly False. Oxford: Oxford University Press; 2012.
36. Kastrup, B. A paradigm-breaking hypothesis for solving the mind-body problem.
Paranthropology: J Anthrop Appr Paran 2012;3:4-12.

21

 37. Van Lommel P. Non-local consciousness: A concept based on scientific research on
near-death experiences during cardiac arrest. J Consciousness Stud 2013;20:7-48.

22

 Table/Figure Legends

Table 1. Items of verifiable information in the first letter

Table 2. Psychographed letters allegedly authored by JP 

Table 3. Frequency, percentage, and mean scores on the Leak Scale

Table 4. Frequency, percentage, and mean scores on the Fit Scale

Table 5. Types of verifiable information

Table 6. Drop-in communications

Figure 1. Persons’ names with Leak scores < 4

23

Figure 1. Person’s names with LEAK scores < 4


24

Table 1. Items of verifiable information in the first
letter

Item Type of information LEAK* FIT°

Sueli First name 1 2

my grandpa Basso [deceased] Surname and kinship 1 2

Elvira [JP’s aunt, deceased] First name 0 2

I see you collecting remembrances in my bedroom as Event post-death 1 2
if I was going to arrive at any instant
Circumstances of 1 2
an outing death

on Sunday Day of death 1 2

Circumstances of 1 2
I gave a hard time to my friends death
Circumstances of 1 2
they called me death
Circumstances of 1 2
they hugged me death
Circumstances of 1 2
they massaged me death
Circumstances of 1 2
they made me breathe death
Circumstances of 1 2
there were no drugs, we were sober death
Circumstances of 1 2
there were no excesses in the eve [of his death] death
Previous activities of 1 2
I miss my studies and classes the deceased
Previous activities of 1 2
the everyday kiss the deceased

Jair First name 1 2

Notes: *LEAK: (0) Very unlikely leak, (1) Rather unlikely leak, (2) Unlikely leak, (3) Likely
leak, (4) Definite leak;
25

°FIT: (0) No Fit, (1) Fit dependent on interpretation or imprecise fit, (2) Clear and precise fit
Table 2. Psychographed letters allegedly authored by JP

Number Date Published Addressed to N N items of

words verifiable

information

1 15/03/1974 1975 Mother, father, 569 16

and sister

2 30/03/1974 1975 Mother, father, 624 12

sister, and

friend

3 25/08/1974 1975 Mother, father, 1095 7

sister, and

friends

4 16/11/1974 1975 Mother, father, 923 7

sister, and

friend

5 29/03/1975 1975 Mother, father, 807 0

sister, and

friends

6 19/07/1975 1976 Mother, father, 1017 8

and sister

7 01/11/1975 1976 Mother, father, 1174 17

and sister

8 14/02/1976 1976 Sister 647 6

9 22/05/1976 No Mother, father, 437 0

26

 and sister

10 13/11/1976 No Mother, father, 633 9

and sister

11 19/03/1977 No Mother, father, 1207 3

and sister

12 14/01/1978 No Mother, father, 953 11

and sister

13 13/01/1979 No Mother, father, 567 3

and sister

27

 Table 3. Frequency, percentage, and mean scores on the LEAK Scale

Letter N items Very unlikely Rather Unlikely leak Likely leak Definitely Not Mean (SD)
leak unlikely leak leak determined
st
1 Letter 16 1 (6.3%) 15 (93.8%) 0 0 0 0 0.94 (0.2)

Letters 1-12 83 8 (9.6%) 0 45 (54.2%) 2 (2.4%) 13 (15.7%) 15 (18.1%) 2.18 (1.1)

28

Table 4. Frequency, percentage, and mean scores on the FIT Scale*

Type of information N items Clear and Fit dependent on No Fit Mean Fit (SD)
Precise Fit Interpretation
Related to JP or his family/friends 64 64 (100%) 0 0 2.00 (0)

Related to individuals not attending 7 5 (71.4%) 2 (28.6%) 0 1.71 (0.5)

the meeting°

Total 71 69 (97.2%) 2 (2.8%) 0 1.97 (0.2)

* in relation to items with LEAK score < 4

° i.e., ‘drop-in’ communication

Table 5. Types of verifiable information*

Type of information Example Frequency

(%)

Name ‘Irineu Leite da Silva’ 16 (22.5%)

Specific event ‘Suzeley [died]… in a swimming pool’ 14 (19.7%)

Feelings/thoughts ‘I ask you [SP] do not feel so down’ 12 (16.9%)

Circumstances of death ‘…heard some folk screaming’ 10 (14.1%)

Name and kinship ‘cousin Jair’ 8 (11.3%)

Name of place/location ‘Praia Azul’ (location of JP’s death) 4 (5.6%)

Reference to specific interests ‘I no longer want to think about trips to the 2 (2.8%)

moon’

Name of institution ‘Grameiro’ (Spiritist Center in Campinas) 2 (2.8%)

Date ‘He [Irineu] died on June 7th last year’ 2 (2.8%)

Reference to specific ‘Oh, Jair, no one can fix him!’ (typical 1 (1.4%)

vocabulary utterance of JP’s friends)

* conveyed by the 71 items with LEAK score < 4

29

Table 6. Drop-in communications
Name of the Information conveyed Means by which information was verified
deceased
personality
Irineu Leite da First name and double surname SP found a small death notice published 41
Silva Date of death (day, month, and days earlier in a local paper of Campinas (a
year) city 400 km away).
Cemetery where he was buried
First names of mother and father

Joãozinho Alves First name and surname A journalist published a small notice in a local
paper of Campinas asking for information
about ‘Joãozinho Alves’, and his parents
contacted JP’s family.

Aníbal First name A journalist published a small notice in a local

Committed suicide paper of Campinas asking for information
about an ‘Aníbal’ who had recently committed
suicide. On the next day, a woman contacted
JP’s family and said that her only child,
named Aníbal, had committed suicide less
than a year ago.

30