Napoleon's Poisoned Chalice

by Dr Martin Howard

  Archibald Arnott was born in Dumfriesshire in 1772 and was educated in medicine at Edinburgh. He joined the 11th Light Dragoons as Surgeon’s Mate and was promoted to Assistant Surgeon in 1799. Eighteen months later, he was appointed Surgeon in the 20th Regiment of Foot and it was with this unit that he saw active service in many of the theatres of the Napoleonic Wars. His Peninsular Medal, now in the museum of the Royal Regiment of Fusiliers, has ten clasps – Egypt, Maida, Vimeiro, Corunna, Vitoria, Pyrenees, Nivelle, Orthez, Nive and Toulouse. He also gained valuable experience in the ill-fated expedition to the Scheldt where many of the troops died of the notorious ‘Walcheren fever’. Arnott contracted the disease and needed two years of home service to recover. After a posting in Ireland, the 20th departed for St. Helena. At this time he was 46 years old, just three years younger than Napoleon. He was, by all accounts, a well respected army doctor and a kindly and likeable man.2

  Because of his regimental seniority and undoubted experience, Arnott was immediately appointed as Principal Medical Officer on the island. This gave him some authority over the other medical officers and it was logical that Lowe should regard him as a possible solution to the stalemate at Longwood where Verling continued his futile vigil. Arnott was to enter centre stage in the spring of 1821 but he made a brief appearance much earlier. On 20th August 1819, Bertrand wrote to the Governor stating that the Emperor had fallen sick in the night; Lowe responded by sending Arnott to offer his services. He simultaneously assured Verling that this was not meant to interfere with his attendance – Arnott’s visit was ‘one of duty as well of attention’. On arrival at Longwood accompanied by Verling, Montholon directed the two doctors to the Grand Marshal’s house. As Arnott was unable to speak French fluently, Verling explained to Bertrand that his colleague was ‘chef des officiers de santé’ and was offering his medical help.

  The conversation that followed between Arnott and Bertrand is well documented by Verling in his journal and the Frenchman in his diary. Both men are largely in agreement with Arnott’s own version of the discussion, which is contained in an incomplete memorandum written for Lowe’s attention.

  I communicated to him [Bertrand] the object of my visit; that I had been ordered by you [Lowe] to offer my assistance in a medical capacity to General Bonaparte, in concert with Dr Verling, to which Count Bertrand replied that the Emperor was very ill, and that although he [the Emperor] entertained a very high opinion of the English Faculty, yet he had refused to be visited by any British surgeon, unless he would accede to certain conditions. The Count also said that he himself and his family having experienced much attention from Dr Verling, had often recommended the Emperor to see him professionally, but that he had uniformly objected to it, because he [Verling] would not accede to the conditions presented by the Emperor. I then signified to Count Bertrand that I would wish the object of my visit to be conveyed to General Bonaparte, and that I would wait at Longwood to know the result. He said that he would communicate the message to the Emperor, and that Dr Verling and I might call on him again. In about an hour afterwards, Count Bertrand sent for Dr Verling and myself to his house. In this second interview he told me he had been with the Emperor and that he would see me, provided I was authorised to give my opinion in writing, to sign and leave it with him. I replied I would not promise to do that without first communicating with you, that I considered my visit strictly professional, and that as such I would act to the best of my judgment but on no other terms would I visit General Bonaparte. ‘Oh!’ then he replied, ‘you are acting under the influence of the Governor.’

  At the same moment the Count took from his side pocket a written paper from which he read several conditions, the purpose of which tended to absolve the person who should have the medical charge of General Bonaparte from your control, and every other military authority. He put the question to me, if I were authorised and willing to give my assent to those articles. I then consequently told the Count …

  We are deprived of the denouement in Arnott’s own words but he predictably refused the Grand Marshal’s offer. It must be remembered that this was the summer of Stokoe’s trial – the court-martial was only two weeks after this meeting – and it would have been an act of gross stupidity to accept the conditions.

  Bertrand could not have been surprised by the outcome but he was frustrated, launching an uncharacteristic attack on Lowe. Why did the Governor send Arnott when he knew that nothing had changed and that there was no chance of his being acceptable to the Emperor? The Governor, he said, obviously believed Napoleon to be ill or he would not be offering a different doctor. So why did he insult him so much? His officers had orders to harass him and to spy on him so that he had to hide behind closed doors and windows. ‘Are these the medicines and remedies that the Governor sends to the sick? What barbarity!’

  Verling called on Montholon who was astonished at the Grand Marshal’s offer. He believed Bertrand to have exceeded his authority from Napoleon; the Emperor had only permitted him to say that if the conditions were met, he would then choose a medical man. The Grand Marshal later approached Verling and confessed that he had not meant to offer Arnott the post, simply to establish the conditions which would have to be met by any British doctor. Perhaps the language proved problematic; both doctors interpreted the Frenchman’s words as an offer of the post of the Emperor’s physician. In truth, Bertrand’s intent was academic as there was no chance of the newly arrived doctor being conferred this honour unless either the British authorities or the Emperor was prepared to make an unexpected concession.

  Despite this stuttering introduction to Longwood, Arnott became friendly with the Bertrands and was a frequent visitor to their house. He assisted Verling and Livingstone in the medical management of the Countess and her children. The Grand Marshal also sought his professional advice; in his diary, he writes, ‘If only the Emperor would consult him!’

  With such close relations with the French, it was inevitable that should the Emperor again become seriously ill, Arnott’s name would be called. Thus it was that, in March 1821 – fully eighteen months after Arnott’s first visit to Longwood – Bertrand and Montholon, alarmed at the physical deterioration of their master, became determined that the doctor should be asked to assist Antommarchi.3

  The pressure on the Emperor to see another British doctor had been building over the previous few months. Lowe again formally offered Arnott’s services in December 1820, writing to Montholon that the medical officer would act as he would for an ‘ordinary private patient’. This was not an altruistic gesture. The Governor was unsettled by reports from Bathurst that Napoleon was planning to escape and he was determined to place one of his subordinates in Longwood to improve surveillance of the prisoner. He informed Montholon that unless the Orderly Officer or a British medical officer was allowed access to the Emperor, he would have to resort to force to gain proof of his presence. The Count, motivated by a different sort of anxiety, had been trying to convince Napoleon to see Arnott since November. He told Lutyens that his master had dismissed his entreaties. ‘I shall be better in a few days, there is no danger.’

  Arnott was edging closer to the Emperor. He met with Antommarchi to discuss the symptoms of the illness. Although he had not seen the patient, the British surgeon concluded that the fever described was not dangerous. He was equally confident that the current illness had no connection with the previous liver trouble. Montholon was now so desperate that Napoleon should see Arnott that he begged him on his knees but, as Marchand relates, he met more resistance.

  ‘What need is there to call another doctor? Will he understand my disease any better? If Corvisart or Larrey were here, I should have confidence and some hope; but these ignoramuses know nothing of my illness. A good ride on a horse – that is what I need! Get a book and read me something.

  Undeterred, Montholon and Bertrand continued to press the case for a second opinion and, at the end of March, Napoleon finally cracked. Marchand explains his decision as being du
e to the renewed pleading of his two companions and also to a realisation that his strength was draining away. This sudden change of heart was still a surprise to those close to him; Saint-Denis admits that it was ‘contrary to all their expectations’. It was a not a gracious acceptance.

  ‘Your British doctor,’ the Emperor said to Bertrand, ‘will go and report my condition to that executioner. It will give him far more pleasure to learn of my agony; but afterwards what else can he do even if I agree to see that doctor? Well! It is more for the satisfaction of the people around me than for my own, for I expect to gain nothing from his opinions!’

  Now that they had obtained their objective, the French were quick to brief Arnott as to how he should behave. Bertrand’s diary entry for 31st March:

  General Montholon spoke to Dr Arnott and asked him if he would give his word of honour to look upon himself as a civilian doctor dealing with an ordinary private person? If he would care for the Emperor as though he were a City doctor attending a London merchant? If he would issue no bulletins or spoken or written reports unless he were so authorised, and would first submit a copy for approval. If he would not speak of the patient other than to say ‘he is well or ill’ without specifying anything about the nature of the illness?

  Arnott stated that he was willing to see the Emperor and could be sent for through the intermediary of Bertrand. He asked if he should sign any bulletin together with Antommarchi as this was usual practice in the case of a private patient. Finally, he sought assurance that if, in the Emperor’s bedroom he were to use ‘another form of address’, that the Governor would not be informed.

  The doctor was nervous of Lowe and did not want his every word to be communicated to Plantation House. Although the Governor had authorised the consultations with the Emperor and had not asked for any bulletins, he was unimpressed at the prospect of another British doctor entering Longwood. He had earlier observed to Verling that any change ‘might give rise to intrigues’. Arnott had a meeting with Lowe prior to the consultation and although we are not party to the details of this, we have the surgeon’s comment that the senior officer was both petulant and furious. This was not unusual for Lowe, but Gorrequer’s contemporary diary entry is explicit as to the Governor’s misgivings.

  Mach’s tortuous proceedings, surrounding every proposition of Medico 20th [Arnott] as well as the fair ones from Veritas [Montholon], with all sorts of captious, peevish objections and difficulties, and agreeing to nothing; evidently wishing, if he could, to prevent him being called in, if he could forge any deceitful or plausible (to Big Wigs at Home) pretext. His repeatedly saying that he should not see him in presence of suite, except Ninny [Reade] was present, or Ninny and Cercueil [Brigadier-General John Pine-Coffin] laying great stress upon making Ninny a part.

  Pine-Coffin had succeeded Bingham in command of the troops. Napoleon took a morbid delight in his name. Arnott was not trusted by either his patient or his military superior.4

  It will be helpful to pause and consider the nature of Napoleon’s disease. An understanding of the illness is important in telling the story of Arnott, a man who was to be haunted by a misdiagnosis. Prior to his demise on St. Helena, Napoleon had enjoyed good health. His iron constitution sustained him through the extremes of the scorching heat of the Egyptian expedition and the piercing cold of the retreat from Moscow. Various authors have tried to attach a number of diagnoses to the Emperor but these do not stand close scrutiny. Segur, in his classic monograph of the Russian campaign, says that Napoleon only ever complained of rheumatism and of dysuria – an irritation on passing urine which may have been caused by bladder stones.

  Following Napoleon’s arrival on the island of his exile, it is possible to divide the history of his health into three phases. The first of these, comfortably the longest, extends from October 1815 to September 1820. During his first two years at Longwood he was mostly well. However, in September 1817 and January 1819 he complained of severe pain in the upper abdomen, pain in the right shoulder blade, nausea and headache. O’Meara was in attendance during the first of these attacks and Stokoe paid his five visits during the second. From this point on, he never regained full health. His physicians noted vomiting, pallor, and constipation alternating with diarrhoea. He also suffered from sporadic episodes of fever, chills, jaundice and darkening of the urine – symptoms which suggest a possible liver problem.

  Concerns regarding the Emperor’s health increased in the summer of 1820 but the second phase of his symptoms can be best defined as being between October 1820 and February 1821. During this period, Antommarchi was on duty and Napoleon’s health greatly deteriorated. Montholon wrote to his wife, ‘The illness of the Emperor has definitely worsened. His pulse is weak; his gums, lips and nails are colourless.’ Every effort was a Herculean task. He complained of persistent abdominal pain, difficulty in swallowing, constipation, night sweats and fever. He was profoundly weak, had no appetite and was losing weight. Madame Bertrand told Assistant Surgeon George Rutledge of the 20th Foot that whilst the others were eating, Napoleon chewed small pieces of under-done meat which he spat out after extracting the juice.

  The final phase was the two months preceding the Emperor’s death, March to May 1821. Antommarchi remained on the scene and Arnott gave his assistance during the last 35 days. The dying man continued to suffer severe abdominal pain, vomiting and fever. The sweating was now so profuse that he had to change his clothes several times every night. His mental state was also in decline; he was intermittently confused and often mistook Arnott for Stokoe or O’Meara. It took a monumental effort and a transient remission in the worst symptoms for the Emperor to dictate his will to Montholon and Marchand. April was marked by a grim accentuation of all the symptoms and, on the 27th he had a ‘coffee ground vomit’ and passed black tarry stools. This, accompanied by an increase in his pulse, was consistent with bleeding from the stomach or the upper part of the gut. In the final week he became first delirious and then comatose. Napoleon remained in this condition until eleven minutes to six on the evening of 5th May 1821, when he died.5

  More clues as to the cause of the disease and death can be gleaned from the post-mortem performed on 6th May by Antommarchi. Seven British doctors, including Arnott, supervised the autopsy. Napoleon’s body was extremely pale. On internal examination, the crucial findings were in the stomach. This was filled with dark material resembling coffee grounds, a sure indication that there had been a sizeable bleed. This was very likely the immediate cause of death. On closer inspection of the wall of the stomach there was a very large ulcer with hard, irregular borders. Separate from this lesion was a second smaller ulcer with adhesions to the liver. There was also some altered blood in the colon, part of the large bowel. The other organs were mostly normal although there was disagreement between those present with regard to the size and state of the liver and there were a few enlarged hardened lymph glands around the stomach.

  As we will never have the opportunity to take a biopsy from Napoleon’s stomach for pathological analysis we have to rely on the combination of his clinical symptoms and the eyewitness observations at the post-mortem to postulate a diagnosis. It is almost certain that he had cancer of the stomach (gastric carcinoma). The nature of the ulcerated lesion in the stomach wall described by Antommarchi corresponds very well with the classical appearance of gastric cancer. A team of pathologists from Dallas have meticulously compared the tumour described in Napoleon’s post-mortem report with modern digital images of stomach cancer and have concluded that it was a particular ulcerating subtype of the disease and that the enlarged lymph glands were proof of spread beyond the organ itself. Even with modern surgical treatment, the prognosis of this type of tumour (stage IIIA) is dismal with less than half of patients still alive one year after diagnosis.

  Why did the Emperor contract this devastating disease? We now know that there are a number of factors which can predispose to the development of gastric cancer. It is possible that he suffered from a chroni
c infection of the stomach (Helicobacter Pylori) that caused both an earlier benign peptic ulcer and ultimately increased his risk of cancer. The separate smaller ulcer found at post-mortem was most likely benign – it was stuck to the liver and was probably the cause of the earlier attacks of abdominal pain. The duration of Napoleon’s symptoms is overly long for stomach cancer alone and the presence of two sorts of disease in the stomach, both benign and malignant, would explain this. His risk of gastric cancer may also have been increased by a high salt intake. His diet was that of a veteran of the Grande Armée and was rich in salt-preserved foods with a dearth of fruit and fresh vegetables.

  Certain types of gastric cancer have a familial predisposition; members of the family carry a genetic abnormality which significantly increases their chance of developing the disease. This may have been the case for the Bonapartes. Napoleon’s father, Charles Bonaparte, died at only 39 years. At post-mortem, he was found to have a tumour the size of a large potato in the lower part of his stomach. Other members of the family did not have autopsies performed and the causes of their deaths are therefore uncertain.6

  Napoleon had a longstanding infection of the stomach, which caused first a benign ulcer and then a fatal cancer. A massive bleed from the tumour hastened his death. This version of events is supported by most doctors who have studied the St. Helena literature but we must also briefly consider the poisoning hypothesis which has its fervent advocates and which has received much publicity. Supporters of this theory claim that Napoleon was deliberately poisoned with arsenic; Montholon is generally held to be the guilty party. The evidence for such a murder is entirely circumstantial and the objective medical information available makes poisoning a very unlikely cause of death. It must be admitted that high levels of arsenic have been found in samples of Napoleon’s hair from St. Helena. This has been presented as scientific proof that arsenic was the cause of his demise. In fact, the hair sample analyses – there have been several performed – are inconclusive. Arsenic and many other potentially dangerous compounds may have entered the hair in a number of innocent ways. Its presence, even in elevated amounts, is neither proof of poisoning nor conclusive evidence that arsenic was the lethal agent. The metal was in widespread use in the early nineteenth century, being a common constituent of drugs and of everyday effects. The wallpaper in the Emperor’s study at Longwood contained significant amounts. Chronic environmental exposure is the likely cause of the laboratory results. This interpretation is strongly supported by a study on sequential hair samples performed in 2004 in Munich. The scientists analysed not only Napoleon’s hairs from St. Helena but also hair samples taken seven years earlier during his incarceration on Elba. All the samples contained equally elevated levels of arsenic in addition to eighteen other metallic elements.