G.G. MEYNELL

A GUIDE TO A DATABASE FOR JOHN LOCKE'S MEDICAL NOTEBOOKS AND MEDICAL READING (revised 8 September 2002)

Medicine and its related sciences were a life-long interest of Locke's, and medical topics are prominent in his numerous common-place books and amongst the titles of his library. However, a comparison of the notebooks and library shows, as might be expected, that he read many books not listed in his library catalogue and that many books that he owned are not mentioned in his notebooks.

For background, see, for example, H.R.F.Bourne, The life of John Locke (London, 1876). K.Dewhurst, John Locke (1632-1704), physician and philosopher: a medical biography (London, 1963). P.Long, A summary catalogue of the Lovelace Collection of the papers of John Locke in the Bodleian Library (Oxford, 1959). P.Long, "The Mellon donation of additional manuscripts of John Locke from the Lovelace Collection". Bodleian Library Record (1962-67), vol.7, pp.185-193. J.Harrison and P.Laslett, The library of John Locke, 1st ed. (Oxford, 1965); 2nd ed. (Oxford, 1971). R.Hall & R.Woolhouse, 80 years of Locke scholarship (Edinburgh, 1983); see The Locke Newsletter or its successor Locke Studies for annual supplements. G.G.Meynell, "A database for John Locke's medical notebooks and medical reading". Medical History (1997), vol.42, pp.473-486; "Locke and alchemy: his notes on Basilius Valentinus and Andreas Cellarius". Locke Studies (2002), vol.2, pp.177-197.

The accompanying database {link} now contains 1,084 records after correcting for the errata described below, and provides a provisional guide to the contents of Locke's medical notebooks. It is not to be regarded as definitive.The files are broadly the same as those on a disc deposited in the Wellcome Institure for the History of Medicine (see Meynell, 1997, note 9) and have now been re-run on a more recent programme. They may be copied without consulting the author who would be grateful to be told of major changes that are subsequently made.

The database arose from a collection of working notes which accumulated over several years and, as such, almost certainly contains inconsistencies and errors that would not be expected in a catalogue proper. Nevertheless, it allows searches to be made far more simply and quickly than would a collection of written notes. It is possible, for example, to select almost instantaneously all the notebooks in which Locke mentions a given title; or all the titles mentioned in a given notebook; or all the books that Locke lent; and so on. Such selections require some knowledge of how the database is designed and this follows here.

The data can be used in two ways (cp.the notes on the Académie des sciences:{link}
1. By using the original database file, BOOKS.DBF on the disc at the Wellcome Library. It was created by the DOS-based programme included in the office suite, LotusWorks ver.1.0 (date, 1990); and Lotus terminology is used here. It can equally well be used with the presentday database, Approach, in Lotus SmartSuite, Millenium Edition (1998), based on Windows 95/98.
Note that the indexes would need to be rebuilt.
2. The alternative is to import the file, ALC_BOOKS.TXT given here (derived by exporting the data from the database as an ASCII file), into an unrelated database programme -- provided the structure of the records is the same. In that file, each record occupies one line and ends with a hard carriage return. Each field ends with a hash (#) as delimiter.
If necessary, the data can also be read from ALC_BOOKS.TXT as a conventional ASCII file, using a word-processor.

To be imported, a new database empty of data should be created by the new database programme with the same fields in the same order as that shown below. Note also that the length of each field can be the same or longer than shown below, but not shorter, or data may be lost. Also additional fields can be added to the end of the above record as desired. The data of ALC_BOOKS.TXT can then be imported into the new database.

RECORD STRUCTURE; FIELDS
Each title found in a notebook occupies one 'record' (the equivalent of one filing card).
Each record here contains 9 'fields' (subheadings) which differ in length, as measured by the total number of characters plus spaces. No field is empty: if there are no data, it starts with a hyphen. Note that the position of a word (or part of a word) in a field does not affect recovery of the record, provided the search is set up correctly (see SEARCHING below).

The nine fields are:

1. CATNO (6: i.e. this field occupies a maximum of six characters or spaces). This is the catalogue number allotted by Harrison & Laslett, The Library Catalogue of John Locke (see above). To ensure that the numbers are sorted into correct numerical order by the computer, blank spaces are filled by '0' (zero) so as to give a maximum of 4 digits. Thus, the published catalogue number, '59', appears as '0059'; but '1014' appears as '1014'.
2. CL (3). Each record was classified as either 'B' (botany), 'C' (chemistry), 'M' (medicine), 'T' (travel), 'O' (miscellaneous), 'K' (amongst alchemical books catalogued by Harrison & Laslett, 2nd ed., p.292), or 'A' (in Adam McLean's List of Authors of Alchemical Books, see www.levity.com/alchemy/authors.html). Some records could obviously have been classified differently (e.g. as 'M' rather than 'C') but this field provides a tentative guide to the subject-matter.
3. DATE (8), as given in the imprint. When Locke gives only the title and the book has only one edition, the dating is straightforward. Otherwise, the date has to rely on Locke's notes which can be unclear.
4. AUTHOR (30). Each name is given as the first four letters only without punctuation. These are sufficient in practice and avoid confusion due to suffixes, e.g. Bartholin, Bartholinus; or Blasius, Blasii. When a name can take several forms, more than one is given: e.g. Cart, Desc; Soha, Zoha; Abul-Pharajio, Grighor (HL.0005).
5. TITLE (30). Here again, a word is given by its first four letters only without punctuation. As pointed out elsewhere, Locke made his notes for his own convenience and the key word(s) which he selected from titles provide many examples, both in his notebooks and in his library catalogue. Although he often used the opening words of a title like 'Praxis Barbett' by Barbette (MS.Locke f.19, p.345), he could choose words placed deep within a long title, as with 'Pomona' which is annexed to Evelyn's 'Sylva' (MS.Locke f.19, p.129): the title field reads, Sylv tree Pomo.
6. PLACE (12), the place of publication taken from the imprint. In general, this is also given as the first 4 letters but note the following:
  Lugd = Lugduni, or Lyons.
  Lug Bat = Lugduni Batavorum, or Leiden.
7. LOCAT (80), location in Locke's notebooks etc. This field often contains several entries, each of which contains the delimiter, $.
A typical entry in this field might therefore read: Bx$f.10 T$f.17 C$d.11, VEe.
(In notebook d.11, each page in divided into columns which Locke labelled by alphabetical letters, both capitals and lower case, as shown here).

The $ sign serves to divide each record into right-hand and left-hand portions:

NOTE:
DL and Fr. often refer to the same date and the same notebook but quote different passages: e.g. Fr, p.251, has entries for Bartholin and Turner which are absent from DL, p.149. On the other hand, DL, p.67, has a citation to Quercitan dated 19.04.1675 that is missing from Fr, p.231.
In the database, DL is generally given by the date (e.g. '$DL:1675.04.19') whereas Fr is generally given by the shelf-mark of the MS.

1. Locke's note of the total number of pages in a book (still very useful in identifying editions). He overlined the number which is shown here between two obliques thus: /425/.
   
2. dates of successive editions, shown by the last digits preceded by an apostrophe. Thus, '1643' is given as '43.
   
3. references to printed bibliographies shown as follows:
   

   16:4121. This is entry 4121 in R.J.Durling, compiler. A catalogue of sixteenth century printed books in the National Library of Medicine. Bethesda, Md. 1967.
   17:.... P.Krivatsy, compiler. A catalogue of seventeenth century printed books in the National Library of Medicine. (Bethesda, Md. 1989).
   18:.... J.B.Blake, compiler. A short title catalogue of eighteenth printed books in the National Library of Medicine. (Bethesda, Md. 1979).
   A&G. A.F.Allison & V.F.Goldsmith. Titles of English books. Two vols. (London, 1977).
   BL. British Library catalogue of printed books.
   BM(NH). Catalogue of the books, manuscripts, maps and drawings in the British Museum (Natural History). Eight vols. (For the Trustees, London, 1903-. repr. 1964).
   B&T. A.L.J.Bayle & A.J.Thillaye. Biographie médicale par ordre chronologique. Two vol., (Paris, 1855; repr. Amsterdam, 1967).
   Celli. B.B.Celli. Bibliografia Hippocratica. (Caracas, 1984).
   Duveen. Bibliotheca alchemica et chemica. An annotated catalogue of printed books on alchemy, chemistry and cognate subjects in the Library of Denis I.Duveen. (London, 1949: repr. Utrecht, 1986).
   Fult. J.F.Fulton. A bibliography of the Honourable Robert Boyle. 2nd ed., (Oxford, 1961).
   H&L. J.Harrison & P.Laslett, The library of John Locke. 2nd ed., (Oxford, 19710.
   Hirsch. A.Hirsch, Biographisches Lexikon der hervorragenden Aertze... Six volumes, (Wien & Leipzig, 1884).
   Pritzel. G.A.Pritzel, Thesaurus literaturae botanicae. New ed. (Berlin, 1871: repr., Milan, 1950).
   RCN. The 'Record Control Number' allotted to an individual record. See, e.g., the BL or NLM databases.
   RCP. Royal College of Physicians of London. Catalogue of the Library. (For the College, London, 1912).
   RSM. Catalogue of the Library of the Royal Medical and Chirurgical Society of London. Three vols. (For the Society, London, 1879). Now in the Royal Society of Medicine, London.
   Wing. D.Wing. Short-title catalogue of books printed in England...1641-1700. 2nd ed. (1994-).
   

   9. MEMO (30) Miscellaneous.

   INDEXES; RECORD SEQUENCES
   As each record is created, the database automatically gives it a unique 'record number'. Thus, record no.539 was the 539th record created; no.1 was the first; and no.1081 was the last. To convert this sequence into an alphabetical sequence based on, for example, the author's name, the collection of records has first to be sorted by the database programme to produce an 'index file'. This can be indexed on a single field like AUTHOR or DATE according to the entries in that field. Alternatively, a single index can be based on two or more fields which operate in succession. Thus, an index might sort first on PLACE and then on DATE. The selected records will first be grouped by PLACE and then, for each group, will be ordered by DATE. Whatever the criteria used for the index file , the underlying collection of records in the database as a whole is left undisturbed: only the index file is different.

   NOTE:
   For each index file, it is essential to check if case-sensitivity is set as 'present' or 'absent'. If 'present', then searching for 'hipp' will not select 'HIPP' or 'Hipp', and so on. If 'absent', any of the three spellings will be selected.
   The field, LOCAT, is designed to be used with an index that is case-Sensitive. However, indexes for the other fields should be case-INsensitive.

   SEARCHING
   Searching may be noticeably faster if the records are initially sorted unindexed (i.e. arranged in order of their record numbers) and the records so selected then arranged by an index.
   In all searches except for empty fields, it is essential to set up the search so that it will detect the search term regardless of whether it is at the beginning or is within a field. If 'Desc' alone is made the search term for AUTHOR, only a record with 'Desc' at the beginning of the AUTHOR field will be detected; but if the search term is entered as *Desc* (to use the Lotus convention), it will be selected whatever its position in the AUTHOR field.
   In a search for an empty field (empty, that is, except for a possible 'OS'; see 'Oak Spring' above), remember that all such fields begin with a hyphen and should be selected by searching for '-', not '*-*' as used for a hyphen within a field.
   In the LOCAT field, the positions of 'T', 'Bd' etc. relative to the $ delimiter differs from record to record. To select them, the wild-card character, ?, standing for any single character is used.
   

   Thus:
   to select for T$, search for 'T$'
   ............. TC$, search for 'T?$'
   ............. TBd$ or TLt$ or TBt$, search for 'T??$'
   ............. TCLt$ etc., search for 'T???$'.
   

   ERRATA

   Five duplicate records were deleted after passing the proofs. The numbers in column 2 of Table 1 should read:

   165 (not 166), 619 (not 622) and 64 (not 65). The total number of records therefore becomes 1,081 (not 1,085).