The user wants me to proofread OCR output of a historical document about Hong Kong police pensions. The text appears to be a table with columns: Name of Pensioner, Amount of Pension in Sterling per annum, Amount of Pension in Dollars per annum, Authority under which the Pension was granted, Date from which the Pension has been paid, Amount of Emolument when last employed in Public Service, Present Age of Pensioner, Cause of Retirement, and Remarks.
The OCR output is messy with broken lines, misaligned columns, and garbled text. I need to reconstruct the table properly in Markdown format.
Let me analyze the data:
First, I see headers:
Then there's "(L 21)" and "POLICE PENSIONS." which might be a page reference or title.
The data rows seem to be:
Let me parse more carefully. The OCR shows:
"Kalla Singh,
72.00
6th Aug., 1909,
216.00
Elu Deen,
Ahmed Khan,....
82.00 82.00
Allah Deen,
46.50
Sunt Singh,.......
Chang Shu-tseung,
Kabu Singh,
Jewar Singh,
Call Singh,
87.06
172.61
78.77
52.70
6th Sept., 1909. Do. 1st Oct., 1909.
5th Jan., 1910.
24th Jan., 1910. 1st Feb., 1910. 1st Mar, 1910.
246.00 246.00 186,00
600.00 216.00 186.00
63.79
Butchen Singh,
87.47
Thomas ledge,
82 10 0
1st April, 1910. 15th May, 1910. 18th May, 1910.
216.00 | 59 246.00
£180
Kishen Singh,..
63.10
1st Sept., 1910.
186.00
Albert Wilson,
2×8.00
William Robertson,
831.00
1st Jan. 1911. 10th June, 1911,
864.00 1,662.00
Amar Singh,
115,00
"Bu lioi,
180,00
Hussein Shab,.
114.00
Chu Ping,
81,00
Makbool Shah,
65.65
Kesur Singh,
62.95
Kadar Bux,..
52.70
Lall Singh,
62.00
Muuga Khan,
60.06
Bikh Mahomei,
52.70
Ordinance No. 11 of 1900,
11th Aug., 1911. 14th Aug., 1911. 19th Nov., 1911. 1st Jan, 1912. 1st March, 1912.
300,00
270.00
342.00
180.00
210.00
1st April, 1912.
186.00
1st April, 1912.
186.00
1st June, 1912.
186.00
16th July, 1912.
216,00
¦
16th Ang., 1912.
186.00
Abmel Deon, Alto Mohamed,
55.80
1- April, 1913.
186,00
47.12
Do.
186.00
Nathy Khan, ....
49.60
Do.
Henry George Baker,
1,178,57 |
10th April, 1913.
146,00 2,016.00
Abdollah,
74.64
16th July, 1913.
Mohr Khan,
55.80
John Lenaghan,
42 0 0
Gunda Singh,
74.78
1st Oct., 1913. 24th Oct., 1913. 1 Mar. 1914.
216.(X)
186,00
ļ
37
£145
61
216.00
Jhaude Khan,
136.50
Kam Elahi.
95.86
Alfred George Dymond
676,80
MontagneGustave Atlee 39 15 10
1st May, 1914.
1st June, 1914.
6 June, 1914. 22nd Ang., 1914.
360,00
% 887 8 285 668 # ER 27822 2 3 2 2 3 985* S KEF 5
63
On Expiration of Term of Service.
62
"
60
61
Medical Certificate.
270,00❘ 65 On Expiration of
Term of Service.
61
"
65
"I
Medical Certificate.
#
67
62
On Expiration of Term of Service.
55
Medical Cer-
tificate.
J
On Expiration of Term of Service.
13
*
69
+
62
91
73
55
63
58
Medical Cer- tificate.
On Expiration of Terms of Service.
Medical Cer- tificate.
On Expiration of Term of Service.
Medical Cer- tificate.
60
56
51
13
"
74
آن
On Expiration of
Term of Service.
Medical Certificate.
"
#!
71
On Expiration of
Ter of Service.
Medical Certificate.
246.00
71
On Expiration of
Term of Service.
1,632.00
£135
Albert Collett,
*99,60
Mohamed Bux,
62.00
23rd May, 1914. 5th Dec., 1914.
2,076.00
196.00
Ratta Singh,
62.00
Pakber Singh,
1st Jan., 1915.
Do.
186.00
62.00
1×6.00
33 OF 65
68
55
Medical Certificate.
61
"7
71
On Expiration of Ten of Service.
67
"
67
15
REMARK.
Died 23rd August, 1936."
This is very messy. I need to reconstruct the table rows. It appears the OCR has read columns vertically or in wrong order. Let me try to identify each pensioner's record.
From the context, this is a list of police pensioners with their pension amounts in sterling and dollars, authority, date, last emolument, age, cause of retirement, and remarks.
I notice some entries have "Ordinance No. 11 of 1900" which is likely the authority for many.
Also "Do." means ditto (same as above).
Let me try to parse row by row. The first few:
Then Butchen Singh - 87.47
Thomas ledge - 82 10 0 (this looks like sterling: £82 10s 0d)
Dates: 1st April 1910, 15th May 1910, 18th May 1910
Emoluments: 216.00, 59? 246.00? £180?
This is extremely messy. The OCR has completely jumbled the columns. I need to make a best effort to reconstruct a plausible table.
Given the instruction: "If data is clearly in table format, reconstruct the table using Markdown table syntax." But the data is so garbled that perfect reconstruction is impossible. However, I must output something in Markdown table format.
I'll try to create a table with the columns as per headers, and fill in what I can, using ... for missing/unclear data.
But the user said: "Preserve original content — do not add or remove substantive words. Do not rephrase, rewrite, or change the author's style, tone, or word choice. Only correct what is clearly wrong."
And "Missing text — if a word or phrase is clearly missing due to OCR damage, insert ... in its place."
So I should not invent data. But I must output a Markdown table. Perhaps I can output the raw text as a table with each row as a pensioner, but the columns are not aligned.
Alternatively, maybe the OCR output is actually a list of lines that correspond to rows, but the columns are separated by spaces. However, the OCR has line breaks in wrong places.
Let me look at the original image? I don't have it. Only the OCR text.
Given the difficulty, I'll attempt to parse the text into a structured table by grouping related entries. But this is speculative.
Perhaps the best approach is to present the data as a cleaned-up list of pensioners with their details as they appear, in a table with columns: Name, Sterling Pension, Dollar Pension, Authority, Date, Last Emolument, Age, Cause, Remarks.
I'll try to extract each pensioner's name and associated numbers.
Let me go through the text sequentially and assign columns based on patterns.
The headers:
Now, the OCR text after headers:
"(L 21) POLICE PENSIONS." - likely a page header.
Then the data starts.
I see "Kalla Singh, 72.00 6th Aug., 1909, 216.00" - This could be: Name: Kalla Singh, Dollar pension: 72.00, Date: 6th Aug 1909, Last emolument: 216.00. Sterling pension missing? Authority missing? Age missing? Cause missing?
Then "Elu Deen, Ahmed Khan,.... 82.00 82.00" - Two names? Or "Elu Deen" and "Ahmed Khan" each with 82.00?
"Allah Deen, 46.50" - Name: Allah Deen, Dollar pension: 46.50?
"Sunt Singh,....... Chang Shu-tseung, Kabu Singh, Jewar Singh, Call Singh, 87.06 172.61 78.77 52.70" - Multiple names then numbers.
Then dates: "6th Sept., 1909. Do. 1st Oct., 1909. 5th Jan., 1910. 24th Jan., 1910. 1st Feb., 1910. 1st Mar, 1910." - six dates.
Then emoluments: "246.00 246.00 186,00 600.00 216.00 186.00" - six amounts.
Then "63.79" - maybe age for one of them?
"Butchen Singh, 87.47" - name and dollar pension?
"Thomas ledge, 82 10 0" - name and sterling pension (£82 10s 0d).
Dates: "1st April, 1910. 15th May, 1910. 18th May, 1910." - three dates.
Emoluments: "216.00 | 59 246.00 £180" - messy.
"Kishen Singh,.. 63.10 1st Sept., 1910. 186.00" - name, dollar pension, date, emolument.
"Albert Wilson, 2×8.00" - name, dollar pension? 2×8.00 = 16? Or 28.00?
"William Robertson, 831.00 1st Jan. 1911. 10th June, 1911, 864.00 1,662.00" - name, dollar pension? date, date, emoluments?
"Amar Singh, 115,00 "Bu lioi, 180,00" - two names?
"Hussein Shab,. 114.00 Chu Ping, 81,00 Makbool Shah, 65.65 Kesur Singh, 62.95 Kadar Bux,.. 52.70 Lall Singh, 62.00 Muuga Khan, 60.06 Bikh Mahomei, 52.70" - list of names with dollar pensions.
Then "Ordinance No. 11 of 1900," - authority.
Dates: "11th Aug., 1911. 14th Aug., 1911. 19th Nov., 1911. 1st Jan, 1912. 1st March, 1912." - five dates.
Emoluments: "300,00 270.00 342.00 180.00 210.00" - five amounts.
Then "1st April, 1912. 186.00 1st April, 1912. 186.00 1st June, 1912. 186.00 16th July, 1912. 216,00 16th Ang., 1912. 186.00" - dates and emoluments.
"Abmel Deon, Alto Mohamed, 55.80 1- April, 1913. 186,00 47.12 Do. 186.00 Nathy Khan, .... 49.60 Do." - names, dollar pensions, dates, emoluments.
"Henry George Baker, 1,178,57 | 10th April, 1913. 146,00 2,016.00" - name, sterling? date, dollar pension? emolument?
"Abdollah, 74.64 16th July, 1913. Mohr Khan, 55.80 John Lenaghan, 42 0 0 Gunda Singh, 74.78 1st Oct., 1913. 24th Oct., 1913. 1 Mar. 1914. 216.(X) 186,00 ļ 37 £145 61 216.00" - messy.
"Jhaude Khan, 136.50 Kam Elahi. 95.86 Alfred George Dymond 676,80 MontagneGustave Atlee 39 15 10 1st May, 1914. 1st June, 1914. 6 June, 1914. 22nd Ang., 1914. 360,00 % 887 8 285 668 # ER 27822 2 3 2 2 3 985 S KEF 5 63 On Expiration of Term of Service. 62 " 60 61 Medical Certificate. 270,00❘ 65 On Expiration of Term of Service. 61 " 65 "I Medical Certificate. # 67 62 On Expiration of Term of Service. 55 Medical Cer- tificate. J On Expiration of Term of Service. 13 69 + 62 91 73 55 63 58 Medical Cer- tificate. On Expiration of Terms of Service. Medical Cer- tificate. On Expiration of Term of Service. Medical Cer- tificate. 60 56 51 13 " 74 آن On Expiration of Term of Service. Medical Certificate. " #! 71 On Expiration of Ter of Service. Medical Certificate. 246.00 71 On Expiration of Term of Service. 1,632.00 £135 Albert Collett, *99,60 Mohamed Bux, 62.00 23rd May, 1914. 5th Dec., 1914. 2,076.00 196.00 Ratta Singh, 62.00 Pakber Singh, 1st Jan., 1915. Do. 186.00 62.00 1×6.00 33 OF 65 68 55 Medical Certificate. 61 "7 71 On Expiration of Ten of Service. 67 " 67 15 REMARK. Died 23rd August, 1936."
This is extremely garbled. The OCR has mixed up columns, and there are many numbers that don't align.
Given the instructions, I must output a Markdown table. I think the best I can do is to create a table with the headers and then each row as a pensioner, but I'll have to leave many cells as ... or with the raw fragmented data.
But the user said: "Do not add or remove substantive words. Do not rephrase, rewrite, or change the author's style, tone, or word choice. Only correct what is clearly wrong."
So perhaps I should not try to restructure into a table if the original is not clearly a table in the OCR? But the headers indicate a table.
The instruction: "Tables — if data is clearly in table format, reconstruct the table using Markdown table syntax." The data is clearly a table format originally, but OCR has destroyed the structure. I should attempt to reconstruct.
Given the difficulty, I'll produce a table with the headers and then list each pensioner as a row with the information I can parse, using ... for missing.
But I need to be careful not to invent data. I'll try to group the data into rows based on the sequence.
Let me attempt to list pensioners in order as they appear:
That's 40 pensioners.
Now for each, I need to assign the columns. But the data is interleaved.
Maybe the original table had multiple columns and the OCR read it column by column? For example, all names first, then all sterling pensions, then all dollar pensions, etc. But the OCR text doesn't show that pattern.
Alternatively, the OCR might have read the table row by row but with line breaks in wrong places.
Given the time, I'll produce a Markdown table with the headers and then for each pensioner, I'll put the name and then ... for other columns, but that seems unhelpful.
Perhaps I can output the raw text as a pre-formatted block? But the instruction says to use Markdown table syntax for tabular data.
I think I need to make a best-effort reconstruction. I'll create a table with the columns and fill in what I can from the text, noting uncertainties.
But the user said: "Only correct what is clearly wrong." So I should not guess.
Maybe the appropriate output is to present the text as a series of paragraphs? But the headers indicate a table.
Let me re-read the prompt: "You are a professional editor specializing in proofreading OCR output of historical records relating to Hong Kong." The primary task: "Reconstruct Sentences and Paragraphs" but also "Tables — if data is clearly in table format, reconstruct the table using Markdown table syntax."
The OCR output is not sentences and paragraphs; it's a table. So I should reconstruct the table.
Given the extreme garbling, I might need to output a table with the headers and then each row as a single cell with the raw line? But that's not a proper table.
I'll try to parse the data into rows by using the dates as delimiters. Each pensioner has a "Date from which the Pension has been paid". There are many dates in the text. Let's list all dates:
That's 32 dates. But there are 40 names. Some pensioners might share dates (Do. means ditto). Also some dates might be for the same pensioner? Unlikely.
Maybe each row has a date. So there should be as many rows as dates. But there are more names than dates.
Let's count names again: I listed 40. But some might be not pensioners but part of other columns? For example, "Ordinance No. 11 of 1900" is authority. "On Expiration of Term of Service" is cause. "Medical Certificate" is cause. "REMARK. Died 23rd August, 1936." is remark.
The numbers like 63, 62, 60, 61, 65, 67, 55, 13, 69, 91, 73, 55, 63, 58, 60, 56, 51, 13, 74, 71, 15 are likely ages.
The text has a block: "63 On Expiration of Term of Service. 62 " 60 61 Medical Certificate. 270,00❘ 65 On Expiration of Term of Service. 61 " 65 "I Medical Certificate. # 67 62 On Expiration of Term of Service. 55 Medical Cer- tificate. J On Expiration of Term of Service. 13 * 69 + 62 91 73 55 63 58 Medical Cer- tificate. On Expiration of Terms of Service. Medical Cer- tificate. On Expiration of Term of Service. Medical Cer- tificate. 60 56 51 13 " 74 آن On Expiration of Term of Service. Medical Certificate. " #! 71 On Expiration of Ter of Service. Medical Certificate. 246.00 71 On Expiration of Term of Service. 1,632.00 £135"
This looks like a column of ages and causes. The numbers 63, 62, 60, 61, 65, 61, 65, 67, 62, 55, 13, 69, 62, 91, 73, 55, 63, 58, 60, 56, 51, 13, 74, 71, 71 are ages. The causes are "On Expiration of Term of Service" or "Medical Certificate". The "270,00" and "246.00" and "1,632.00" and "£135" might be emoluments or sterling pensions.
This is too messy. Given the constraints, I think the best is to output a Markdown table with the headers and then note that the OCR output is too garbled to reconstruct accurately, but the instruction says to output only the proofread text in Markdown. I must output something.
Perhaps I can output the text as a cleaned-up version preserving the original line breaks but fixing hyphenation and spacing, and then add a table for the parts that are clearly tabular? But the whole thing is a table.
I'll attempt to create a table with the headers and then each row as a pensioner with the data I can extract, using ... for missing. I'll base it on the sequence of names and the dates/emoluments that follow.
Let me try to pair names with the nearest date and emolument.
From the start:
"Kalla Singh, 72.00 6th Aug., 1909, 216.00" -> Name: Kalla Singh, Dollar pension: 72.00, Date: 6th Aug 1909, Last emolument: 216.00. Sterling pension? Authority? Age? Cause? Remarks?
"Elu Deen, Ahmed Khan,.... 82.00 82.00" -> Two pensioners? Elu Deen and Ahmed Khan, each with dollar pension 82.00? No dates yet.
"Allah Deen, 46.50" -> Allah Deen, dollar pension 46.50.
"Sunt Singh,....... Chang Shu-tseung, Kabu Singh, Jewar Singh, Call Singh, 87.06 172.61 78.77 52.70" -> Five names, four dollar amounts? Maybe Sunt Singh, Chang Shu-tseung, Kabu Singh, Jewar Singh, Call Singh. But only four numbers. Perhaps the first number belongs to Sunt Singh? 87.06, 172.61, 78.77, 52.70 for the last four? But there are five names.
Then dates: "6th Sept., 1909. Do. 1st Oct., 1909. 5th Jan., 1910. 24th Jan., 1910. 1st Feb., 1910. 1st Mar, 1910." That's six dates. Could correspond to six pensioners: maybe Elu Deen, Ahmed Khan, Allah Deen, Sunt Singh, Chang Shu-tseung, Kabu Singh? But there are more names.
Then emoluments: "246.00 246.00 186,00 600.00 216.00 186.00" six amounts.
Then "63.79" maybe age for one.
"Butchen Singh, 87.47" -> Butchen Singh, dollar pension 87.47.
"Thomas ledge, 82 10 0" -> Thomas Ledge, sterling pension £82 10s 0d.
Dates: "1st April, 1910. 15th May, 1910. 18th May, 1910." three dates.
Emoluments: "216.00 | 59 246.00 £180" messy.
"Kishen Singh,.. 63.10 1st Sept., 1910. 186.00" -> Kishen Singh, dollar pension 63.10, date 1st Sept 1910, emolument 186.00.
"Albert Wilson, 2×8.00" -> Albert Wilson, dollar pension 28.00? (2×8.00)
"William Robertson, 831.00 1st Jan. 1911. 10th June, 1911, 864.00 1,662.00" -> William Robertson, dollar pension 831.00? dates 1st Jan 1911 and 10th June 1911? emoluments 864.00 and 1,662.00?
"Amar Singh, 115,00 "Bu lioi, 180,00" -> Amar Singh 115.00, Bu Lioi 180.00.
"Hussein Shab,. 114.00 Chu Ping, 81,00 Makbool Shah, 65.65 Kesur Singh, 62.95 Kadar Bux,.. 52.70 Lall Singh, 62.00 Muuga Khan, 60.06 Bikh Mahomei, 52.70" -> list of names with dollar pensions.
Then "Ordinance No. 11 of 1900," authority.
Dates: "11th Aug., 1911. 14th Aug., 1911. 19th Nov., 1911. 1st Jan, 1912. 1st March, 1912." five dates.
Emoluments: "300,00 270.00 342.00 180.00 210.00" five amounts.
Then "1st April, 1912. 186.00 1st April, 1912. 186.00 1st June, 1912. 186.00 16th July, 1912. 216,00 16th Ang., 1912. 186.00" four dates with emoluments.
"Abmel Deon, Alto Mohamed, 55.80 1- April, 1913. 186,00 47.12 Do. 186.00 Nathy Khan, .... 49.60 Do." -> Abmel Deon, Alto Mohamed, Nathy Khan with dollar pensions, dates, emoluments.
"Henry George Baker, 1,178,57 | 10th April, 1913. 146,00 2,016.00" -> Henry George Baker, sterling pension £1,178.57? date 10th April 1913, dollar pension 146.00? emolument 2,016.00.
"Abdollah, 74.64 16th July, 1913. Mohr Khan, 55.80 John Lenaghan, 42 0 0 Gunda Singh, 74.78 1st Oct., 1913. 24th Oct., 1913. 1 Mar. 1914. 216.(X) 186,00 ļ 37 £145 61 216.00" -> messy.
"Jhaude Khan, 136.50 Kam Elahi. 95.86 Alfred George Dymond 676,80 MontagneGustave Atlee 39 15 10 1st May, 1914. 1st June, 1914. 6 June, 1914. 22nd Ang., 1914. 360,00 % 887 8 285 668 # ER 27822 2 3 2 2 3 985 S KEF 5 63 On Expiration of Term of Service. 62 " 60 61 Medical Certificate. 270,00❘ 65 On Expiration of Term of Service. 61 " 65 "I Medical Certificate. # 67 62 On Expiration of Term of Service. 55 Medical Cer- tificate. J On Expiration of Term of Service. 13 69 + 62 91 73 55 63 58 Medical Cer- tificate. On Expiration of Terms of Service. Medical Cer- tificate. On Expiration of Term of Service. Medical Cer- tificate. 60 56 51 13 " 74 آن On Expiration of Term of Service. Medical Certificate. " #! 71 On Expiration of Ter of Service. Medical Certificate. 246.00 71 On Expiration of Term of Service. 1,632.00 £135 Albert Collett, *99,60 Mohamed Bux, 62.00 23rd May, 1914. 5th Dec., 1914. 2,076.00 196.00 Ratta Singh, 62.00 Pakber Singh, 1st Jan., 1915. Do. 186.00 62.00 1×6.00 33 OF 65 68 55 Medical Certificate. 61 "7 71 On Expiration of Ten of Service. 67 " 67 15 REMARK. Died 23rd August, 1936."
This is hopeless to parse accurately.
Given the instructions, I think I should output a Markdown table with the headers and then for each pensioner, a row with the name and as much data as I can confidently assign, leaving others as ... But I must not invent.
Perhaps the best is to output the text as a cleaned-up version with paragraphs? But the instruction says to use Markdown table for tabular data.
I'll compromise: I'll create a table with the headers, and then include the raw data as a single cell per row? No.
Maybe I can output the text as a pre-formatted block (using code fence) but the instruction says "Do not wrap your response in code fences." So no code fences.
I'll output a Markdown table with the headers and then one row that says "Data too garbled to reconstruct accurately" but that's not the original content.
The user might expect me to do my best to reconstruct. As an AI, I should attempt to parse the data into a structured table.
Let me try a different approach: The OCR might have read the table column by column. For example, first column: names. Second column: sterling pensions. Third column: dollar pensions. Fourth: authority. Fifth: date. Sixth: emolument. Seventh: age. Eighth: cause. Ninth: remarks.
If I can separate the text into columns by looking at the order of appearance. But the text is linear.
Maybe the original image had two pages? The "(L 21)" might be page 21.
Given the time, I'll produce a table with the headers and then list each pensioner name in a row, with other columns as ... except where I can clearly see a value associated.
But that would be adding content (the names) but not the other data.
I think I need to accept that I cannot perfectly reconstruct, but I must output something. I'll create a table with the headers and then for each distinct name I identified, a row with the name and the dollar pension amount if it appears right after the name, and the date if it appears near, etc. But it's speculative.
Given the instruction "Only correct what is clearly wrong." I should not guess associations.
Perhaps the correct approach is to output the text as a series of paragraphs, each representing a row, with the columns separated by tabs or something? But the instruction says to use Markdown table syntax.
I'll write a Markdown table with the headers and then a single row with the entire OCR text in the first cell? That seems wrong.
Let me read the prompt again: "Return the proofread text in standard Markdown: - Use #, ##, ### for headers found in the original. - Use bold for labels, titles, and emphasized text (e.g., RESTRICTED, CONFIDENTIAL, MEMORANDUM). - Use Markdown table syntax (| col | col |) to reconstruct tabular data. - Wrap file references and codes in backticks. - Do not wrap your response in code fences. - Do not output HTML tags. - Do not add any commentary, notes, or explanations."
So I must output the proofread text. The original text is a table. I must reconstruct the table.
I will attempt to reconstruct the table by making reasonable assumptions. Since this is a historical record, maybe the data is known. But I don't have external knowledge.
I'll do my best to align the data. Let me try to parse the text into a CSV-like structure by using the fact that the OCR might have inserted line breaks at column boundaries.
I'll write a script in my mind to parse. But as a language model, I can simulate.
Let me copy the OCR text into a string and try to split by lines. The user provided the text as a single block. I'll split by newline.
The user's message has line breaks. Let me see the raw input: It's a single message with line breaks. I'll treat each line as a line.
But the user's message in the prompt is the OCR output. It has line breaks. Let me count lines.
I'll copy the OCR text from the user message (the part after "Name of Pensioner. ..." up to "Died 23rd August, 1936.").
I'll split by newline and see.
But in this reasoning environment, I can't run code. I'll do manually.
The text lines (as displayed):
2.
5.
7.
9.
11.
Now, this is a line-by-line representation. The original table likely had multiple columns. The OCR has read it line by line, but the columns are not aligned.
Given the headers, there are 9 columns. The first few lines are headers split across lines.
Lines 1-4: "Name of Pensioner." "Amount of the Pension" "in Sterling per annum." -> that's one header: "Name of Pensioner", "Amount of the Pension in Sterling per annum."
Line 6: "( L 21 )" maybe page number.
Line 8: "POLICE PENSIONS." title.
Line 10: "Amount of the Pension in Dollars per annum."
Line 12-14: "Authority under which the Pension" "WHE" "grante)." -> "Authority under which the Pension was granted."
Line 15-17: "Date from which" "the Pension" "Las been paid." -> "Date from which the Pension has been paid."
Line 18-19: "Amount of Emolument when last" "employed in Public Service."
Line 20-21: "Present Age of" "Pensioner."
Line 22: "Cause of Retirement."
Line 23-26: "£" "8. d." "C." "C." -> likely subheaders: £, s., d., and two columns for dollars? "C." might be for cents? Or "C." for "Currency"? Actually "C." might be for "Dollars" and "Cents"? But there are two "C.".
Maybe the table has columns: Name, £, s., d., Dollars, Cents, Authority, Date, Emolument, Age, Cause, Remarks.
But the headers show "Amount of the Pension in Sterling per annum." and "Amount of the Pension in Dollars per annum." So two amount columns: one in sterling (£ s. d.), one in dollars (maybe with cents). Then Authority, Date, Emolument, Age, Cause, Remarks.
The "£ 8. d. C. C." might be the subheaders for the sterling and dollar columns: £, s., d., $, c.? But "C." appears twice.
Anyway, the data lines start at line 27.
Now, the data seems to be arranged in a way that each pensioner's information spans multiple lines. But the OCR has broken each column into separate lines? For example, line 27: "Kalla Singh," (name). Line 28: "72.00" (maybe dollar amount). Line 29: "6th Aug., 1909," (date). Line 30: "216.00" (emolument). Then line 31: "Elu Deen," (next name). Line 32: "Ahmed Khan,...." (another name?). Line 33: "82.00 82.00" (two dollar amounts). Line 34: "Allah Deen," (name). Line 35: "46.50" (dollar). Line 36: "Sunt Singh,......." (name). Line 37: "Chang Shu-tseung," (name). Line 38: "Kabu Singh," (name). Line 39: "Jewar Singh," (name). Line 40: "Call Singh," (name). Line 41: "87.06" (dollar). Line 42: "172.61" (dollar). Line 43: "78.77" (dollar). Line 44: "52.70" (dollar). Line 45: "6th Sept., 1909. Do. 1st Oct., 1909." (dates). Line 46: "5th Jan., 1910." (date). Line 47: "24th Jan., 1910. 1st Feb., 1910. 1st Mar, 1910." (dates). Line 48: "246.00 246.00 186,00" (emoluments). Line 49: "600.00 216.00 186.00" (emoluments). Line 50: "63.79" (age?). Line 51: "Butchen Singh," (name). Line 52: "87.47" (dollar). Line 53: "Thomas ledge," (name). Line 54: "82 10 0" (sterling: £82 10s 0d). Line 55: "1st April, 1910. 15th May, 1910. 18th May, 1910." (dates). Line 56: "216.00 | 59 246.00" (emoluments and age?). Line 57: "£180" (sterling pension?). Line 58: "Kishen Singh,.." (name). Line 59: "63.10" (dollar). Line 60: "1st Sept., 1910." (date). Line 61: "186.00" (emolument). Line 62: "Albert Wilson," (name). Line 63: "2×8.00" (dollar? 28.00). Line 64: "William Robertson," (name). Line 65: "831.00" (dollar). Line 66: "1st Jan. 1911. 10th June, 1911," (dates). Line 67: "864.00 1,662.00" (emoluments). Line 68: "Amar Singh," (name). Line 69: "115,00" (dollar). Line 70: ""Bu lioi," (name). Line 71: "180,00" (dollar). Line 72: "Hussein Shab,." (name). Line 73: "114.00" (dollar). Line 74: "Chu Ping," (name). Line 75: "81,00" (dollar). Line 76: "Makbool Shah," (name). Line 77: "65.65" (dollar). Line 78: "Kesur Singh," (name). Line 79: "62.95" (dollar). Line 80: "Kadar Bux,.." (name). Line 81: "52.70" (dollar). Line 82: "Lall Singh," (name). Line 83: "62.00" (dollar). Line 84: "Muuga Khan," (name). Line 85: "60.06" (dollar). Line 86: "Bikh Mahomei," (name). Line 87: "52.70" (dollar). Line 88: "Ordinance No. 11 of 1900," (authority). Line 89: "11th Aug., 1911. 14th Aug., 1911. 19th Nov., 1911. 1st Jan, 1912. 1st March, 1912." (dates). Line 90: "300,00" (emolument). Line 91: "270.00" (emolument). Line 92: "342.00" (emolument). Line 93: "180.00" (emolument). Line 94: "210.00" (emolument). Line 95: "1st April, 1912." (date). Line 96: "186.00" (emolument). Line 97: "1st April, 1912." (date). Line 98: "186.00" (emolument). Line 99: "1st June, 1912." (date). Line 100: "186.00" (emolument). Line 101: "16th July, 1912." (date). Line 102: "216,00" (emolument). Line 103: "¦" (artifact). Line 104: "16th Ang., 1912." (date). Line 105: "186.00" (emolument). Line 106: "Abmel Deon, Alto Mohamed," (names). Line 107: "55.80" (dollar). Line 108: "1- April, 1913." (date). Line 109: "186,00" (emolument). Line 110: "47.12" (age?). Line 111: "Do." (ditto for authority). Line 112: "186.00" (emolument). Line 113: "Nathy Khan, ...." (name). Line 114: "49.60" (dollar). Line 115: "Do." (ditto). Line 116: "Henry George Baker," (name). Line 117: "1,178,57 |" (sterling? £1,178.57). Line 118: "10th April, 1913." (date). Line 119: "146,00 2,016.00" (dollar pension? emolument?). Line 120: "Abdollah," (name). Line 121: "74.64" (dollar). Line 122: "16th July, 1913." (date). Line 123: "Mohr Khan," (name). Line 124: "55.80" (dollar). Line 125: "John Lenaghan," (name). Line 126: "42 0 0" (sterling £42). Line 127: "Gunda Singh," (name). Line 128: "74.78" (dollar). Line 129: "1st Oct., 1913. 24th Oct., 1913. 1 Mar. 1914." (dates). Line 130: "216.(X)" (emolument?). Line 131: "186,00" (emolument). Line 132: "ļ" (artifact). Line 133: "37" (age). Line 134: "£145" (sterling pension). Line 135: "61" (age). Line 136: "216.00" (emolument). Line 137: "Jhaude Khan," (name). Line 138: "136.50" (dollar). Line 139: "Kam Elahi." (name). Line 140: "95.86" (dollar). Line 141: "Alfred George Dymond" (name). Line 142: "676,80" (sterling? £676.80). Line 143: "MontagneGustave Atlee 39 15 10" (name and sterling? £39 15s 10d). Line 144: "1st May, 1914." (date). Line 145: "1st June, 1914." (date). Line 146: "6 June, 1914. 22nd Ang., 1914." (dates). Line 147: "360,00" (emolument). Line 148: "% 887 8 285 668 # ER 27822 2 3 2 2 3 985 S KEF 5" (garbage). Line 149: "63" (age). Line 150: "On Expiration of Term of Service." (cause). Line 151: "62" (age). Line 152: """ (ditto). Line 153: "60" (age). Line 154: "61" (age). Line 155: "Medical Certificate." (cause). Line 156: "270,00❘ 65 On Expiration of" (emolument and age and cause). Line 157: "Term of Service." (cause). Line 158: "61" (age). Line 159: """ (ditto). Line 160: "65" (age). Line 161: ""I" (artifact). Line 162: "Medical Certificate." (cause). Line 163: "#" (artifact). Line 164: "67" (age). Line 165: "62" (age). Line 166: "On Expiration of Term of Service." (cause). Line 167: "55" (age). Line 168: "Medical Cer-" (cause). Line 169: "tificate." (cause). Line 170: "J" (artifact). Line 171: "On Expiration of Term of Service." (cause). Line 172: "13" (age). Line 173: "" (artifact). Line 174: "69" (age). Line 175: "+" (artifact). Line 176: "62" (age). Line 177: "91" (age). Line 178: "73" (age). Line 179: "55" (age). Line 180: "63" (age). Line 181: "58" (age). Line 182: "Medical Cer- tificate." (cause). Line 183: "On Expiration of Terms of Service." (cause). Line 184: "Medical Cer- tificate." (cause). Line 185: "On Expiration of Term of Service." (cause). Line 186: "Medical Cer- tificate." (cause). Line 187: "60" (age). Line 188: "56" (age). Line 189: "51" (age). Line 190: "13" (age). Line 191: """ (ditto). Line 192: "74" (age). Line 193: "آن" (Arabic? artifact). Line 194: "On Expiration of" (cause). Line 195: "Term of Service." (cause). Line 196: "Medical Certificate." (cause). Line 197: """ (ditto). Line 198: "#!" (artifact). Line 199: "71" (age). Line 200: "On Expiration of" (cause). Line 201: "Ter of Service." (cause). Line 202: "Medical Certificate." (cause). Line 203: "246.00" (emolument). Line 204: "71" (age). Line 205: "On Expiration of" (cause). Line 206: "Term of Service." (cause). Line 207: "1,632.00" (emolument). Line 208: "£135" (sterling pension). Line 209: "Albert Collett," (name). Line 210: "*99,60" (dollar? £99.60?). Line 211: "Mohamed Bux," (name). Line 212: "62.00" (dollar). Line 213: "23rd May, 1914. 5th Dec., 1914." (dates). Line 214: "2,076.00" (emolument). Line 215: "196.00" (emolument). Line 216: "Ratta Singh," (name). Line 217: "62.00" (dollar). Line 218: "Pakber Singh," (name). Line 219: "1st Jan., 1915." (date). Line 220: "Do." (ditto). Line 221: "186.00" (emolument). Line 222: "62.00" (dollar). Line 223: "1×6.00" (dollar). Line 224: "33 OF 65" (artifact). Line 225: "68" (age). Line 226: "55" (age). Line 227: "Medical Certificate." (cause). Line 228: "61" (age). Line 229: ""7" (artifact). Line 230: "71" (age). Line 231: "On Expiration of Ten of Service." (cause). Line 232: "67" (age). Line 233: """ (ditto). Line 234: "67" (age). Line 235: "15" (age). Line 236: "REMARK." (header). Line 237: "Died 23rd August, 1936." (remark).
This is a line-by-line breakdown. The data is essentially a list of fields in column-major order? It seems the OCR read the table column by column: first all names, then all sterling pensions, then all dollar pensions, then authorities, then dates, then emoluments, then ages, then causes, then remarks. But the lines are interleaved.
Actually, looking at the lines, after the headers, we have a block of names (lines 27,31,32,34,36,37,38,39,40,51,53,58,62,64,68,70,72,74,76,78,80,82,84,86,106,113,116,120,123,125,127,137,139,141,143,209,211,216,218). That's many names.
Then dollar amounts appear at lines 28,33,35,41,42,43,44,52,59,63,65,69,71,73,75,77,79,81,83,85,87,107,114,121,124,128,138,140,142,143?,210,212,217,222,223.
Sterling amounts at lines 54,57,117,142,143,208.
Dates at lines 29,45,46,47,55,60,66,89,95,97,99,101,104,108,118,122,129,144,145,146,213,219.
Emoluments at lines 30,48,49,56,61,67,90,91,92,93,94,96,98,100,102,105,109,112,119,130,131,136,147,148?,203,207,214,215,221.
Ages at lines 50,56?,110,133,135,149,151,153,154,156,158,160,164,165,167,172,174,176,177,178,179,180,181,187,188,189,190,192,199,204,225,226,228,230,232,234,235.
Causes at lines 150,152,155,156,157,162,166,168,169,171,182,183,184,185,186,194,195,196,200,201,202,205,206,227,231,233.
Remarks at line 237.
Authority at line 88,111,115.
So the OCR has read the table column by column! That is, it read the first column (names) top to bottom, then second column (sterling?), but not exactly.
Actually, the order of lines suggests the OCR read the table in a serpentine or column-wise manner. But the lines are in the order they appear in the OCR output. The user gave us the OCR output as a text block. That text block is the result of OCR reading the image, likely left-to-right, top-to-bottom, but the table might have been multi-column on the page? Or the OCR just output the text in reading order of the image, which for a table might be row by row but with line breaks at cell boundaries.
Given the line-by-line analysis, it's clear that the OCR output is a linear sequence of text fragments that correspond to table cells in some order. To reconstruct the table, we need to map each cell to its row and column.
This is a complex puzzle. However, as an AI, I can attempt to reconstruct by assuming the table has a fixed number of rows and columns, and the OCR read it row by row but each cell on a new line? But the number of lines per row varies.
Given the difficulty, perhaps the expected output is to present the data as a cleaned-up table with the headers and then each row as a pensioner with the information we can deduce. But the instruction says "Do not add or remove substantive words." So I must not invent associations.
Maybe the correct approach is to output the text as a series of paragraphs, each representing a row, with the columns separated by | ? But the instruction says to use Markdown table syntax.
I think I will output a Markdown table with the headers, and then for each pensioner name that appears, create a row with the name and leave other columns blank (or ...). But that would be adding rows for each name, which might be correct if each name is a pensioner.
But there are 40 names. The table likely has 40 rows.
I'll create a table with 9 columns as per headers. I'll fill in the name column with each distinct name in order of appearance. For other columns, I'll try to match the data that appears in proximity in the line sequence.
But the line sequence is not row-major. However, the names appear in the order of the rows? Probably the names are listed in the first column in order. So the first name is Kalla Singh, second Elu Deen, third Ahmed Khan, fourth Allah Deen, fifth Sunt Singh, sixth Chang Shu-tseung, seventh Kabu Singh, eighth Jewar Singh, ninth Call Singh, tenth Butchen Singh, eleventh Thomas Ledge, twelfth Kishen Singh, thirteenth Albert Wilson, fourteenth William Robertson, fifteenth Amar Singh, sixteenth Bu Lioi, seventeenth Hussein Shab, eighteenth Chu Ping, nineteenth Makbool Shah, twentieth Kesur Singh, twenty-first Kadar Bux, twenty-second Lall Singh, twenty-third Muuga Khan, twenty-fourth Bikh Mahomei, twenty-fifth Abmel Deon, twenty-sixth Alto Mohamed, twenty-seventh Nathy Khan, twenty-eighth Henry George Baker, twenty-ninth Abdollah, thirtieth Mohr Khan, thirty-first John Lenaghan, thirty-second Gunda Singh, thirty-third Jhaude Khan, thirty-fourth Kam Elahi, thirty-fifth Alfred George Dymond, thirty-sixth Montagne Gustave Atlee, thirty-seventh Albert Collett, thirty-eighth Mohamed Bux, thirty-ninth Ratta Singh, fortieth Pakber Singh.
That's 40 names.
Now, the other columns: We have 40 entries for each column? Let's count dollar amounts: there are many dollar amounts. But we need to assign each to a row.
The dollar amounts appear in the line sequence after some names. For example, after Kalla Singh (line27), line28 is 72.00. After Elu Deen and Ahmed Khan (lines31,32), line33 has 82.00 82.00. After Allah Deen (line34), line35 46.50. After a block of five names (lines36-40), lines41-44 have four dollar amounts. That suggests maybe the first of those five names (Sunt Singh) doesn't have a dollar amount? Or the dollar amounts correspond to the last four names.
Then after Butchen Singh (line51), line52 87.47. After Thomas Ledge (line53), line54 is sterling 82 10 0, not dollar. After Kishen Singh (line58), line59 63.10. After Albert Wilson (line62), line63 2×8.00. After William Robertson (line64), line65 831.00. After Amar Singh (line68), line69 115,00. After Bu Lioi (line70), line71 180,00. After Hussein Shab (line72), line73 114.00. After Chu Ping (line74), line75 81,00. After Makbool Shah (line76), line77 65.65. After Kesur Singh (line78), line79 62.95. After Kadar Bux (line80), line81 52.70. After Lall Singh (line82), line83 62.00. After Muuga Khan (line84), line85 60.06. After Bikh Mahomei (line86), line87 52.70.
That's 24 dollar amounts for the first 24 names? Let's count: Kalla Singh (1), Elu Deen (2), Ahmed Khan (3), Allah Deen (4), then four for the next four of the five? Actually names 5-9: Sunt Singh, Chang Shu-tseung, Kabu Singh, Jewar Singh, Call Singh. Dollar amounts: 87.06, 172.61, 78.77, 52.70 (four amounts). So maybe one of them missing. Then Butchen Singh (10) 87.47. Thomas Ledge (11) has sterling not dollar. Kishen Singh (12) 63.10. Albert Wilson (13) 28.00. William Robertson (14) 831.00. Amar Singh (15) 115.00. Bu Lioi (16) 180.00. Hussein Shab (17) 114.00. Chu Ping (18) 81.00. Makbool Shah (19) 65.65. Kesur Singh (20) 62.95. Kadar Bux (21) 52.70. Lall Singh (22) 62.00. Muuga Khan (23) 60.06. Bikh Mahomei (24) 52.70.
That's 24 names with dollar amounts (though Thomas Ledge has sterling). Then after that, we have authority "Ordinance No. 11 of 1900" at line88. Then dates at line89 (five dates). Then emoluments at lines90-94 (five amounts). Then more dates and emoluments for subsequent names.
This suggests the table might have multiple sections. The first section (up to Bikh Mahomei) might be one group, then a new group starting with authority Ordinance No. 11 of 1900 for the next set.
The next names: Abmel Deon, Alto Mohamed (line106), Nathy Khan (line113), Henry George Baker (line116), Abdollah (line120), Mohr Khan (line123), John Lenaghan (line125), Gunda Singh (line127), Jhaude Khan (line137), Kam Elahi (line139), Alfred George Dymond (line141), Montagne Gustave Atlee (line143), Albert Collett (line209), Mohamed Bux (line211), Ratta Singh (line216), Pakber Singh (line218). That's 16 names.
But there are also names
Name of Pensioner.
Amount of the Pension
in Sterling per annum.
( L 21 )
POLICE PENSIONS.
Amount of the Pension in Dollars per annum.
Authority under which the Pension
WHE
grante).
Date from which
the Pension
Las been paid.
Amount of Emolument when last
employed in Public Service.
Present Age of
Pensioner.
Cause of Retirement.
£
C.
C.
Kalla Singh,
72.00
6th Aug., 1909,
216.00
Elu Deen,
Ahmed Khan,....
82.00 82.00
Allah Deen,
46.50
Sunt Singh,.......
Chang Shu-tseung,
Kabu Singh,
Jewar Singh,
Call Singh,
87.06
172.61
78.77
52.70
6th Sept., 1909. Do. 1st Oct., 1909.
5th Jan., 1910.
24th Jan., 1910. 1st Feb., 1910. 1st Mar, 1910.
246.00 246.00 186,00
600.00 216.00 186.00
63.79
Butchen Singh,
87.47
Thomas ledge,
82 10 0
1st April, 1910. 15th May, 1910. 18th May, 1910.
216.00 | 59 246.00
£180
Kishen Singh,..
63.10
1st Sept., 1910.
186.00
Albert Wilson,
2×8.00
William Robertson,
831.00
1st Jan. 1911. 10th June, 1911,
864.00 1,662.00
Amar Singh,
115,00
"Bu lioi,
180,00
Hussein Shab,.
114.00
Chu Ping,
81,00
Makbool Shah,
65.65
Kesur Singh,
62.95
Kadar Bux,..
52.70
Lall Singh,
62.00
Muuga Khan,
60.06
Bikh Mahomei,
52.70
Ordinance No. 11 of 1900,
11th Aug., 1911. 14th Aug., 1911. 19th Nov., 1911. 1st Jan, 1912. 1st March, 1912.
300,00
270.00
342.00
180.00
210.00
1st April, 1912.
186.00
1st April, 1912.
186.00
1st June, 1912.
186.00
16th July, 1912.
216,00
¦
16th Ang., 1912.
186.00
Abmel Deon, Alto Mohamed,
55.80
1- April, 1913.
186,00
47.12
Do.
186.00
Nathy Khan, ....
49.60
Do.
Henry George Baker,
1,178,57 |
10th April, 1913.
146,00 2,016.00
Abdollah,
74.64
16th July, 1913.
Mohr Khan,
55.80
John Lenaghan,
42 0 0
Gunda Singh,
74.78
1st Oct., 1913. 24th Oct., 1913. 1 Mar. 1914.
216.(X)
186,00
ļ
37
£145
61
216.00
Jhaude Khan,
136.50
Kam Elahi.
95.86
Alfred George Dymond
676,80
MontagneGustave Atlee 39 15 10
1st May, 1914.
1st June, 1914.
6 June, 1914. 22nd Ang., 1914.
360,00
% 887 8 285 668 # ER 27822 2 3 2 2 3 985* S KEF 5
63
On Expiration of Term of Service.
62
"
60
61
Medical Certificate.
270,00❘ 65 On Expiration of
Term of Service.
61
"
65
"I
Medical Certificate.
#
67
62
On Expiration of Term of Service.
55
Medical Cer-
tificate.
J
On Expiration of Term of Service.
13
*
69
+
62
91
73
55
63
58
Medical Cer- tificate.
On Expiration of Terms of Service.
Medical Cer- tificate.
On Expiration of Term of Service.
Medical Cer- tificate.
60
56
51
13
"
74
آن
On Expiration of
Term of Service.
Medical Certificate.
"
#!
71
On Expiration of
Ter of Service.
Medical Certificate.
246.00
71
On Expiration of
Term of Service.
1,632.00
£135
Albert Collett,
*99,60
Mohamed Bux,
62.00
23rd May, 1914. 5th Dec., 1914.
2,076.00
196.00
Ratta Singh,
62.00
Pakber Singh,
1st Jan., 1915.
Do.
186.00
62.00
1×6.00
33 OF 65
68
55
Medical Certificate.
61
"7
71
On Expiration of Ten of Service.
67
"
67
15
REMARK.
Died 23rd August, 1936.
No comments yet.
Private notes are available after approval.