FOIA 25-254: Two certificates of nonexistence, two documents produced, and a $74,520 itemization copied from an earlier request

Watchdog Alerts

Why isn’t this basic information available to the public?

The office certifies that it keeps no index, spreadsheet or log of when Brady or Giglio materials were served, and in the same letter describes a written policy that requires six law-enforcement agencies to report eight categories of officer-credibility information to it, plus a compliance form for collecting that information. It keeps the intake and not the record of what it did with it.
The $74,520.00 estimate is numerically identical to the estimate the office issued in FOIA 25-244 on October 30, 2025: same $49,680.00 to locate at 8,000 increments, same $24,840.00 to redact at 4,000 increments, same $6.21 rate. 25-254 covers 2022 to the present; 25-244 covered 2020 through 2025. The itemization did not change when the request did.
The itemization bills minutes as if they were quarter-hours. Line 1 states '800 cases @ 10 minutes' and then enters 8,000 increments; the form's own 15-minute rule makes that 533 increments, or $3,309.93. Line 3a states 6,000 files at five minutes and enters 4,000 increments; the same rule makes that 2,000 increments, or $12,420.00.
The cover letter says the office will review approximately 200 cases per year. The itemization's justification line says 1,000 cases per year for six years — the volume and the period from FOIA 25-244, left in place on a request covering roughly four years.
The cover letter estimates 90 days from receipt of the deposit. The itemization states 180 days. The two documents were signed the same day.
The Notice of Denial checks all three reason boxes at once — Exempt from Disclosure, Record Does Not Exist and Redaction — and never says which reason governs which paragraph.
Seven weeks before this response, the office produced thirteen Brady/Giglio guidance documents in FOIA 25-244, including quick reference guides, a manual, timing-of-disclosure sheets and annotated examples. Asked again in paragraph two for 'letters, form templates, or checklists used by prosecutors to determine when disclosure is required,' the office produced two documents and did not mention the thirteen.
The request asked for a public-interest fee waiver under MCL 15.234. The Public Interest section of the county's own itemization form is left entirely blank — no waiver, no reduction, no written reason.

Timing caveat. ProsecutorByKarpel went live in June 2024. This request reaches back to January 1, 2022, so part of the window predates the system, and nothing on this page faults the office for how those earlier files were kept. The point is narrower. Paragraph three asked for any Karpel report or export referencing Brady, Giglio, officer credibility or similar fields. The office certified that none exist, then granted 'the underlying information that would allow for the compilation of the statistics requested' for $74,520.00. If the underlying information can be compiled, the question is what it is stored in and why a query of the system the county bought in 2024 is not the cheaper route. The letter does not describe a search, does not name Karpel, and does not say whether the system was opened. It prices a manual review of case files instead, at a rate and a volume that match the previous request rather than this one.

On October 31, 2025, a request went to the Delta County Prosecutor’s Office asking for something narrow and specific: the records that show when Brady and Giglio material was turned over to defense counsel, the templates the office uses to decide when turnover is required, and any report from the case-management system that tracks officer-credibility fields. It was faxed at 10:48:14 that morning and logged as received on November 3. The office filed a notice to extend on November 6 and answered on November 20, 2025.

The answer was mixed, and the mix is what makes it worth reading. The office granted paragraph two outright and produced two real documents — a five-question Brady/Giglio Compliance Form and a February 15, 2024 memo to six law-enforcement agencies, with an eight-part addendum on officer-credibility information. Then it certified that no Brady/Giglio disclosure log exists, certified that no case-management report or export exists, and priced the “underlying information” at $74,520.00 with a $37,260.00 deposit.

That $74,520.00 is not a new number. It is the same number, on the same county form, with the same handwritten case counts and the same “6 YEARS requested” language, as the itemization the office issued in FOIA 25-244 — dated October 30, 2025, one day before this request arrived, for a different request covering a different period. The invoice was prepared before the request existed.

How to read this page

This page separates six things that are easy to blur together. First, what was asked. Second, what the office actually produced, which is more than in most responses in this archive. Third, the two certificates of nonexistence and the sentence that grants the “underlying information” they supposedly cover. Fourth, the arithmetic, which on this itemization does not work — not by a little, but by a factor of exactly fifteen. Fifth, the fact that the itemization is a reuse of the one issued in 25-244, down to the handwriting. Sixth, the public-interest waiver, which was requested and which the form leaves blank.

Where the record supports criticism, the criticism is made and the document is cited. Where the record is silent, this page says so. The office’s letter is quoted at length because its own words carry the argument better than a summary would. Redacted copies of the response, the denial form, the itemization, the request and the two produced records are linked under Documents below.

Summary

The request was titled “FOIA Request – Brady and Giglio Disclosure Logs and Templates (2022–Present)” and organized in three numbered paragraphs. Paragraph one asked for any index, spreadsheet or log recording when Brady or Giglio materials were served on defense counsel. Paragraph two asked for the letters, form templates or checklists the office uses to determine when disclosure is required. Paragraph three asked for any report or export from ProsecutorByKarpel — the office’s case-management system, which the request named — referencing Brady, Giglio, “officer credibility,” or similar fields.

The office answered each paragraph differently, which is itself unusual in this archive. Most responses from this office apply one disposition to the whole request and then describe it inconsistently. This one distinguishes:

Paragraph one: denied, with a certificate of nonexistence. The office also denied a second time on a prior-production theory, writing that the information “is in the form of Proofs of Service” that were “previously provided to you in response to FOIA Request 25-244,” and that because the requester “ha[s] not provided good cause on why a reproduction of that information is required, the request is DENIED pursuant to the precedent in Densmore v Department of Corrections, 203 Mich 363 (1994).”
Paragraph two: granted. Two records produced: the Brady/Giglio Compliance Form and the February 15, 2024 memo with its addendum.
Paragraph three: granted in part, denied in part. Denied as to the ProsecutorByKarpel reports and exports, with a second certificate. Granted “as to the underlying information that would allow for the compilation of the statistics requested” — for a fee.

The fee is $74,520.00. The office describes the work as reviewing “approximately 200 cases per year,” at “approximately 10 minutes per case, per year” to locate responsive records and “approximately 5 minutes per case” to separate exempt from nonexempt material. It requires a 50 percent deposit and, in the letter, estimates 90 days from receipt of the deposit. The attached itemization says 180 days.

Neither the letter nor the itemization explains why a request for a log — a list of dates on which something happened — requires reading case files at all. If a log exists, it is a document and it can be copied. If it does not exist, that is the answer, and the office gave that answer too. What the office did not explain is how both answers can be true at once, or why the price for the second one is the same price it charged for an entirely different request seven weeks earlier.

What was asked, and how the office answered

The request is dated October 31, 2025. The TelNet fax receipt in the packet records transmission at 10:48:14 a.m. that day, one page, to 906-789-5197. The office’s Notice of Denial records the request as received on 11/03/2025. On November 6, 2025 the office filed a Notice to Extend Response Time, signed that day. The response letter is dated November 20, 2025, on Delta County Prosecutor’s Office letterhead naming Lauren M. Wickman as Prosecuting Attorney and Civil Counsel, Beth Wickwhite as Chief Assistant Prosecutor and Deputy Civil Counsel, and Lauren Mattson as Assistant Prosecutor. It is signed by Emily DeSalvo as FOIA Coordinator, with a contact number of 906-789-5100.

On timing, the office complied, and this page says so here rather than at the end. Five business days from a Monday, November 3 receipt runs to Monday, November 10. The ten-business-day extension available under MCL 15.235(4) runs to November 24. A response dated November 20 sits inside that window, and the extension was noticed in writing on the fourth business day after receipt. There is no timing problem in this packet.

The three paragraphs, as the request framed them:

“(1) Any index, spreadsheet, log, or similar record that documents or tracks when Brady or Giglio materials were disclosed or served upon defense counsel; (2) any letters, form templates, or checklists used by the office to determine when Brady or Giglio disclosure is required; and (3) any reports or exports generated by the ProsecutorByKarpel (PbK) case-management system referencing Brady, Giglio, ‘officer credibility,’ or similar fields.”

The disposition line for paragraph one reads “DENIED,” and the certificate is printed in full:

“Your request is DENIED, as a review of the records maintained by the Delta County Prosecutor’s Office reveals that no ‘[a]ny index, spreadsheet, log, or similar record’ … exist[s]. The undersigned hereby certifies that, after a search for records and to the best of the undersigned’s knowledge and belief, the public records requested do not exist within the records of the public body under the name nor any other name reasonably known to the public body.”

Paragraph three gets the same treatment for the system query, and then a sentence that changes the shape of the response:

“Your request is GRANTED as to the underlying information that would allow for the compilation of the statistics requested. However, given the breadth of your request and the number of records which must be searched and reviewed to respond to such a request, the public body has determined that failing to charge for the costs of the search, examination, review, and deletion and separation of exempt from nonexempt information would result in unreasonably high costs to the County beyond those typically incurred in responding to FOIA requests.”

This is the same structure that appears in FOIA 25-257, issued the same day: a certificate that closes the door, followed immediately by a sentence that opens it, followed by an invoice that closes it again. In 25-257 the pivot sentence was “however, the data related to the request is available.” Here it is “granted as to the underlying information.” The wording differs; the move is identical.

The paper trail

The packet is 29 pages, and it is complete. It contains the response letter, the Notice of Denial, the text of MCL 15.240, a blank FOIA Appeal Form, the county’s written FOIA policies, a seven-page Detailed Cost Itemization, the Fee Appeal Form and the text of MCL 15.240a, a signed extension form, the fax records from both ends, the original request, and the two produced records. A requester who wanted to appeal had everything needed to do so in one envelope.

FOIA 25-254 — documents and dates as they appear in the packet
Date	Document	What it establishes
Oct 31, 2025	Request letter, three numbered paragraphs	Scope: Brady/Giglio disclosure logs, disclosure templates, and PbK reports or exports referencing Brady, Giglio or officer-credibility fields, 2022 to present. Fee waiver requested under MCL 15.234.
Oct 31, 2025, 10:48:14 a.m.	TelNet fax portal receipt	Transmission time to the second; one page; to 906-789-5197.
Nov 3, 2025	Office receipt date	The operative receipt date, stated on the Notice of Denial.
Nov 6, 2025	Notice to Extend Response Time	Ten-business-day extension, filed on the fourth business day after receipt.
Nov 20, 2025	Response letter and Notice of Denial	Paragraph 1 denied twice; paragraph 2 granted; paragraph 3 granted in part and denied in part; the $74,520.00 estimate and 90-day estimate.
Nov 20, 2025	Detailed Cost Itemization, seven pages	$49,680.00 locate, $24,840.00 redact, $74,520.00 subtotal, $37,260.00 deposit, 180-day estimate. Waiver section blank.
Feb 15, 2024	Memo and Addendum, produced	The office’s written Brady/Giglio guidance to six law-enforcement agencies, with eight categories of officer-credibility information agencies must report.
Undated	Brady/Giglio Compliance Form, produced	A five-question form for each law-enforcement witness, with a YES/NO box and an instruction to supply details in a separate document.

One line in the letter matters more than the rest of the chronology. It reads: “Please note this request is nearly identical to the request in FOIA Request 25-214, paragraph 3.” That is the office telling the requester, in its own letter, that this ground has been covered before. The archive does not contain FOIA 25-214. This page describes it only as the office described it and does not speculate about its contents.

Two certificates, and the grant that sits between them

The response contains two certificates of nonexistence. The first covers paragraph one: no index, spreadsheet or log tracking when Brady or Giglio materials were served. The second covers paragraph three: no report or export from ProsecutorByKarpel referencing Brady, Giglio, officer credibility or similar fields. Between them sits the grant of “the underlying information that would allow for the compilation of the statistics requested.”

Read narrowly, all three statements can be defended, and this page will defend them before it criticizes them. “Records” means documents the office already holds. “Underlying information” means the raw material inside case files from which such a document could be assembled. An office that keeps no Brady/Giglio log but does keep the case files can truthfully say both things. That is a coherent position, and a public body is not obliged to create a record it never made.

The problem is what follows from it. If the underlying information exists in the office’s files, then the question the requester actually asked — when was Brady or Giglio material turned over, and to whom — is answerable from records the office holds. The only remaining questions are the form of the answer and who pays for the assembly. The office chose the most expensive available form: a human being reading case files one at a time, at ten minutes to locate and five minutes to redact, across roughly 200 cases per year.

It did not choose the form the request asked for, which was a query against the system the office uses to track exactly this kind of information. And unlike 25-257, where the letter never mentioned Karpel at all, this letter does mention it — it names the system in the certificate, certifies that no report or export from it exists, and then prices the manual alternative without ever saying whether the system was queried, what fields it contains, or whether a query would have returned anything.

There is a third possibility, and it is the one the record cannot rule out: the fields are populated, the query would work, and the manual read is the form of production chosen because it produces a number large enough to end the request. This page does not assert that. It asserts only that the letter does not describe a search of the system, does not describe the system’s fields, and does not explain why the expensive method was chosen over the cheap one. On a $74,520.00 estimate, that explanation is the least a requester could ask for.

The invoice that was written before the request

This is the finding that makes 25-254 different from every other fee dispute in this archive, and it is checkable in about a minute by putting two documents side by side.

The Detailed Cost Itemization attached to 25-254 is dated November 20, 2025. The Detailed Cost Itemization attached to FOIA 25-244 is dated October 30, 2025. The 25-254 request was faxed on October 31, 2025. The invoice now attached to it was prepared, at the latest, the day before the request arrived.

And it is the same invoice. Not similar. The same figures, the same rate, the same increment counts, the same handwritten case populations, and the same period language:

FOIA 25-244 itemization (Oct 30, 2025) against FOIA 25-254 itemization (Nov 20, 2025)
Element	25-244, dated Oct 30, 2025	25-254, dated Nov 20, 2025
Hourly wage with fringe	$24.82	$24.82
Charge per quarter hour	$6.21	$6.21
Line 1, labor to locate	8,000 increments × $6.21 = $49,680.00	8,000 increments × $6.21 = $49,680.00
Line 3a, labor to redact	4,000 increments × $6.21 = $24,840.00	4,000 increments × $6.21 = $24,840.00
Subtotal	$74,520.00	$74,520.00
Deposit (50 percent)	$37,260.00	$37,260.00
Estimated time frame	180 days	180 days
Handwritten locate basis	“reviewing approximately 1000 to 1000 cases/year (6 YEARS requested)”	“reviewing approximately 1000 to 1000 cases/year (6 YEARS requested)”
Handwritten redact basis	“1000 in each request/year for 6 years”	“1000 in each request/year for 6 years”
Period covered by the request	2020 through 2025	2022 to present

The two requests are not the same request. FOIA 25-244 asked for records about the use of a particular set of charging and sentencing tools over a six-year window. FOIA 25-254 asks for Brady/Giglio disclosure logs, templates, and officer-credibility fields over a period that begins in 2022 — roughly four years, not six. The subject matter is different. The paragraph structure is different. The number of years is different.

The price is identical, to the dollar and to the increment. So is the handwriting that explains it. The “6 YEARS requested” line, which was accurate for 25-244’s 2020–2025 window, is simply false for a request covering 2022 to the present. It was not corrected. It was copied.

There are two possible explanations, and only one of them is innocent. The first is clerical: the coordinator attached the wrong itemization, or reused the previous form as a template and did not update the figures. That happens, and it is a documentation failure rather than a decision. The second is that the estimate was never computed from this request at all — that $74,520.00 is the office’s standing price for a broad records request, applied without regard to what was asked. Under either explanation, the number attached to this response is not an estimate of the cost of this request. It is a number that existed before the request did.

Either way, the consequence for the requester is the same. A $74,520.00 demand with a $37,260.00 deposit is not a price a private individual pays to find out whether a prosecutor’s office keeps a disclosure log. It is a number that ends the conversation. And on this record, the office did not have to compute it. It already had it.

The $74,520.00, and the arithmetic that does not survive it

The itemization is a seven-page county form. The figures are set out exactly as printed.

Detailed Cost Itemization, FOIA 25-254, dated November 20, 2025
Line	Item	Basis as printed	Amount
1	Labor cost to locate	8,000 quarter-hour increments × $6.21	$49,680.00
2	Labor cost for copying	—	$0.00
3a	Labor cost to redact	4,000 quarter-hour increments × $6.21	$24,840.00
3b	Contract labor cost to redact	—	$0.00
4	Copying and duplication	—	$0.00
5	Mailing	—	blank
Subtotal fees			$74,520.00
Deposit required (50 percent)			$37,260.00
Estimated time frame			180 days

Now the part that matters. The form’s own rate is $6.21 per quarter hour. That is fifteen minutes. The form’s own handwritten justification supplies the case counts and the minutes per case. Put those three facts together and the printed increment counts cannot be reached.

Printed increments against the increments the form’s own method produces
Step	Line 1, locate	Line 3a, redact
Cases or files, as handwritten on the form	800 cases	6,000 files (1,000 × 6 years)
Minutes per case or file, as handwritten	10	5
Total minutes	8,000	30,000
Minutes in a quarter hour	15	15
Quarter-hour increments this produces	533	2,000
Cost at $6.21 per increment	$3,309.93	$12,420.00
Total on the form’s own method	$15,729.93
Increments printed on the itemization	8,000	4,000
Amount printed on the itemization	$49,680.00	$24,840.00
Total printed	$74,520.00

The printed figures treat minutes as if they were quarter hours. Eight thousand minutes of locating work became 8,000 quarter hours — 2,000 hours — where the form’s own method yields 533 increments, or 133 hours. Four thousand is what the redaction line would be if the per-file time were 10 minutes rather than 5, or if the file count were 12,000 rather than 6,000; neither is what the handwriting says.

There is a second way to see the same defect, and it uses only the letter. The response letter states that “a review of approximately 200 cases per year will be required,” at ten minutes per case per year to locate and five minutes per case to separate exempt material. Take the letter at its word across the four years the request actually covers — 2022, 2023, 2024 and 2025:

The letter’s own volume, priced at the letter’s own rate
Step	Locate	Redact
Cases per year, as stated in the letter	200	200
Years covered by the request (2022 to present)	4	4
Total cases	800	800
Minutes per case	10	5
Total minutes	8,000	4,000
Quarter-hour increments	533	267
Cost at $6.21 per increment	$3,309.93	$1,657.47
Total on the letter’s own figures	$4,967.40

The printed subtotal is $74,520.00. The letter’s own volume, priced at the letter’s own rate and the form’s own increment, comes to roughly $4,967. The ratio between them is exactly fifteen — which is the arithmetic signature of minutes being billed as quarter hours. The estimate is not merely generous. It is off by the precise factor that a units error produces.

Figure 1 · Interactive

Increments printed on the itemization against increments its own method produces

The form bills in quarter-hour increments of $6.21. Its handwritten case counts and per-case minutes produce far fewer increments than it charges for. Select a bar to read the arithmetic.

Locate — printed 8,000 Line 1 as printed: 8,000 quarter-hour increments at $6.21, for $49,680.00. That is 2,000 hours of locating work. The handwritten basis on the same form says 800 cases at 10 minutes each — 8,000 minutes, which is 133 hours.

Locate — 15-minute rule 533 800 cases × 10 minutes = 8,000 minutes. Divided by 15 minutes per increment = 533 increments, or $3,309.93. This is what the form’s own rate and the form’s own case count produce.

Redact — printed 4,000 Line 3a as printed: 4,000 quarter-hour increments at $6.21, for $24,840.00. That is 1,000 hours of redaction. The handwritten basis says 1,000 files in each request year for 6 years at 5 minutes per file — 30,000 minutes, which is 500 hours.

Redact — 15-minute rule 2,000 6,000 files × 5 minutes = 30,000 minutes. Divided by 15 minutes per increment = 2,000 increments, or $12,420.00. Twice the printed figure would still be half the printed amount.

Printed subtotal$74,520

Form’s own method$15,730

Letter’s own volume$4,967

Printed ÷ letter15.0×

Source: Detailed Cost Itemization attached to FOIA 25-254, dated November 20, 2025, pages 9–15 of the response packet, read from a 200 dpi render of the handwritten justification; and the response letter of the same date, which states “approximately 200 cases per year,” “approximately 10 minutes per case, per year” and “approximately 5 minutes per case.” All arithmetic computed from the printed figures.

Figure 1. The itemization charges for 8,000 and 4,000 quarter-hour increments. Its own case counts and per-case minutes produce 533 and 2,000. The printed subtotal is exactly fifteen times what the letter’s own volume yields at the form’s own rate.

Two things should be said in the office’s favor here, because they are true. First, the rate itself is disclosed and reasonable: $24.82 an hour with fringe benefits is a real county wage, and $6.21 per quarter hour is exactly that rate divided by four. Second, the deposit is exactly half the subtotal, as the statute permits. The office did not hide its rate or pad its deposit. It disclosed everything a requester needs to check the work — and the work does not check out.

That is the difference between this itemization and the one in 25-257. The 25-257 arithmetic is internally consistent: 1,200 files a year, five minutes to locate, three minutes to redact, 2,000 and 1,200 increments, $19,840.00. Every line follows from the premise. The 25-254 arithmetic does not follow from its premise, and it does not follow because it was not derived from its premise. It was copied from a form where the same error had already been made.

Two documents, two different volumes, two different deadlines

The letter and the itemization disagree about how much work is involved, and they disagree about how long it will take. Neither disagreement is reconciled anywhere in the packet.

Where the letter and the itemization diverge
Question	Response letter, Nov 20, 2025	Cost itemization, Nov 20, 2025
How many cases per year	“approximately 200 cases per year”	“approximately 1000 to 1000 cases/year (6 YEARS requested)”
How many years	2022 to present — about four	“6 YEARS requested”
Minutes to locate, per case	“approximately 10 minutes per case, per year”	“800 cases @ 10 minutes”
Minutes to redact, per case	“approximately 5 minutes per case”	“1000 in each request/year for 6 years” at 5 minutes per file
Time to deliver after deposit	“approximately 90 days from receipt of the deposit”	180 days

The volume gap is five to one. The letter says 200 cases a year; the form says 1,000. The letter covers roughly four years; the form says six. If the letter’s figure is right, the request involves about 800 cases. If the form’s figure is right, it involves 6,000 files. Both documents are dated November 20, 2025, both were signed or prepared by the same office on the same day, and neither acknowledges the other.

The deadline gap is smaller but cleaner. The letter gives 90 days from receipt of the deposit. The itemization gives 180 days. A requester who pays $37,260.00 is entitled to know which one governs, and the packet does not say. Under MCL 15.235(4) an extension runs ten business days, not ninety or one hundred eighty; both figures are the office’s own production estimates rather than statutory deadlines, but a fee estimate that carries two different completion dates is a fee estimate a requester cannot rely on.

There is one more volume question the packet never answers, and it is the one a requester would ask first. If the request is for a log — a list of dates on which Brady or Giglio material was served — why does answering it require reading case files at all? A log is either a document that exists, in which case it is copied, or it is a document that does not exist, in which case the certificate is the whole answer. The office gave the certificate. It then priced the alternative anyway, as though the requester had asked for the underlying facts rather than the list. On paragraph one, the requester asked for the list.

A denial form that checks all three boxes

Page four of the packet is the state-issued Notice of Denial form. It records request number 25-254, date received 11/03/2025, date of notice 11/20/2025, received via fax, contact Emily DeSalvo at 906-789-5100. In the field labeled “Record(s) You Requested” it says “see attached copy of the original request” — which is at least an improvement on 25-257, where the locate line from the itemization was entered in that field instead of a description of the records.

Three reason boxes are checked, all of them:

Reason 1, Exempt from Disclosure, with the explanation “records previously disclosed pursuant to FOIA Request 25-244.”
Reason 2, Record Does Not Exist, with the standard certificate language.
Reason 3, Redaction, with the subsection field and the description field partly filled.

Those three reasons cannot all be true of the same record. A record that does not exist cannot be exempt because it was previously disclosed, and a record that was previously disclosed exists. A record that is redacted exists. The form is a template designed to be filled in for one disposition per record, and here it records three, without mapping any of them to a paragraph.

The letter does better than the form, and this page credits that. The letter distinguishes the paragraphs: paragraph one denied, paragraph two granted, paragraph three granted in part and denied in part. A reader of the letter can reconstruct which reason applies where. A reader of the form cannot. And the form is the document a requester attaches to an appeal, because it is the document titled “Notice of Denial.”

The Reason 1 explanation deserves its own paragraph, because it is a distinct legal theory and it is the second denial of paragraph one. The office writes that the information requested “is in the form of Proofs of Service” that were “previously provided to you in response to FOIA Request 25-244,” and that because the requester has not shown good cause for a reproduction, the request is denied “pursuant to the precedent in Densmore v Department of Corrections, 203 Mich 363 (1994).”

Two observations, and neither is an accusation. First, the citation is given as 203 Mich 363 (1994). A Michigan Supreme Court volume 203 dates from about 1918; the 1994 Densmore decision is reported at 203 Mich App 363. This page quotes the citation exactly as the office printed it and does not resolve the discrepancy, because resolving it is the office’s job and the underlying proposition — that a public body may decline to supply a second copy of records already furnished — is not controversial. Second, and more substantively: the office says the responsive information was already produced in 25-244. That is a concession that the information exists and has been assembled before. It is difficult to reconcile with a simultaneous certificate that no such record exists, and it is impossible to reconcile with a $74,520.00 estimate for assembling it again.

The waiver section, left blank

The request asked for a public-interest fee waiver under MCL 15.234. On the county’s own itemization form, the Public Interest section is present and completely empty. Neither the box for “all fees are waived” nor the box for “all fees are reduced by ___ percent” is checked. The line for “Subtotal Fees After Waiver” is blank. The indigence section is unchecked. The nonprofit discount section is unchecked. The section for an increased deposit due to previously unpaid fees is blank as well.

This is the quietest move in the packet and, on this record, the most consequential. Declining a waiver with a written reason creates a decision that can be reviewed, appealed, and argued about. Leaving the section blank creates nothing. There is no finding to challenge, no rationale to test, no date on which a decision was made. The full $74,520.00 simply stands, and the requester’s request for a waiver is answered with white space.

MCL 15.234(7) permits a public body to waive or reduce a fee where the request is in the public interest because it is likely to contribute significantly to public understanding of the operations or activities of government and is not primarily in the commercial interest of the requester. Whether a prosecutor’s office keeps a Brady/Giglio disclosure log, and what its written disclosure policy says to the six agencies it prosecutes for, is about as direct a question about the operations of government as FOIA produces. The office did not have to grant the waiver. It did have to say something.

This is now the fourth request in this archive in which a waiver was asked for and the form’s waiver section came back blank. The pattern is consistent enough to name: the one part of the fee form that requires a human being to write down a reason is the part that is consistently left empty.

What the office did produce

Paragraph two was granted in full, and the two documents produced are the most useful things in the packet. They are also, read carefully, the strongest argument against the rest of the response.

The Brady/Giglio Compliance Form. A single page, one per law-enforcement witness, with fields for the witness’s name and the case name, and five questions:

Length of employment with the agency.
Does this witness have any findings of misconduct during the course of their employment? YES / NO. If yes, “in a sperate document, please provide details of the misconduct related to dishonesty or credibility.”
Does this witness have any prior criminal convictions? YES / NO. If yes, provide details in a separate document.
Are you aware of any current or ongoing investigations related to dishonesty, credibility, or criminal activity pending against this witness? YES / NO. If yes, provide details.
Are you aware of any other issues involving dishonesty or credibility for this witness that could be considered impeachment evidence? YES / NO. If yes, provide details.

The form is a screening instrument. It asks an officer, or an agency, to disclose impeachment material about a witness before that witness testifies. It is the intake side of a Brady/Giglio process. What it is not is a record of output — it does not record when material was served, on whom, or in which case.

The February 15, 2024 memo and addendum. Four pages, addressed to Sheriff Ed Oswald of the Delta County Sheriff’s Office, Director John Gudwer of Escanaba Public Safety, Director Ron Robinson of Gladstone Public Safety, F/Lt Gregg Cunningham of the Michigan State Police, D/Lt Tim Sholander of the Upper Peninsula Substance Enforcement Team, and Chief Jeff Hanson of the Hannahville Tribal Police Department. Signed by Lauren M. Wickman as Delta County Prosecuting Attorney. Subject: “Continuing Duty to Disclose Exculpatory and Favorable Impeachment Evidence.”

The memo is a serious document, and it is written well. It explains the “prosecution team” concept, distinguishes Brady material from Giglio material, notes that the law presumes the team knows what any individual member knows, and warns that withholding such evidence “may also have serious adverse personal consequences on members of the prosecution team and their careers.” It sets out five categories of exculpatory evidence and four of impeachment evidence. It is the kind of guidance a prosecutor’s office should send, and the office sent it to six agencies.

The addendum is where the record gets interesting. Titled “Addendum Regarding Law Enforcement Witnesses,” it states that “it is important that our office receive the following information concerning any member of your agency who has served or will serve as an affiant or testify in court,” and then lists eight categories:

The eight categories of officer-credibility information the office asked six agencies to report, February 15, 2024
Item	Information the agency is asked to provide
(a)	Any finding of misconduct that reflects upon the truthfulness or possible bias of the employee, including a finding of lack of candor during a criminal, civil, or administrative inquiry or proceeding.
(b)	Any past or pending criminal charge brought against the employee.
(c)	Any allegation of misconduct bearing upon truthfulness, bias, or integrity that is the subject of a pending investigation.
(d)	Any prior finding by a judge that the employee has testified untruthfully, made a knowingly false statement in writing, engaged in an unlawful search or seizure, illegally obtained a confession, or engaged in other misconduct.
(e)	Any misconduct finding or pending misconduct allegation that casts substantial doubt on the accuracy of evidence the prosecutor intends to rely on, or that might bear significantly on the admissibility of prosecution evidence.
(f)	Failure to follow mandatory protocols with regard to the forensic analysis of evidence.
(g)	Information that may be used to suggest that the officer is biased for or against a particular defendant.
(h)	Information that the witness’s ability to perceive and/or recall the truth is impaired.

Then comes the sentence that ties this document to the rest of the response:

“Please be aware that impeachment information relating to law enforcement witnesses disclosed to our office will not automatically be disclosed to the court or to the defense. Rather, our policy is that such information will be reviewed by the Prosecuting Attorney or Chief Assistant Prosecuting Attorney, who will decide whether further exploration of the facts should be made with the involved agency. No decision about disclosure will be made without consultation with the leadership of the agency involved.”

That is a written policy, in the office’s own words, describing an internal review process with a decision point. Information comes in from six agencies across eight categories. The Prosecuting Attorney or Chief Assistant reviews it. A decision is made about whether to explore further. And then, presumably, a decision is made about whether to disclose.

A process with an intake, a reviewer, a decision point and an output is a process that produces a record. It may produce a memo, a note in a file, an email, or an entry in a case-management system. It does not have to produce a document titled “Brady/Giglio Log.” But somewhere between the agency’s report and the defense’s receipt, something is written down — or the office is asserting that a formal, multi-agency, multi-category review process runs entirely without paper.

That is the tension this packet never resolves. The office produced a memo establishing a disclosure-review process, produced a form for collecting the inputs to that process, certified that no record tracks the outputs of that process, and then priced the “underlying information” at $74,520.00. Each of those four statements is individually defensible. Together they describe an office that has built the machinery of Brady/Giglio compliance and keeps no record of whether it works.

What 25-244 already produced, seven weeks earlier

FOIA 25-244 was answered on October 30, 2025 — twenty-one days before this response, and one day before this request was faxed. In that response the office produced thirteen Brady/Giglio guidance documents, including the same February 15, 2024 memo and the same Compliance Form that appear in this packet.

That matters for two reasons. First, it means paragraph two of this request was, in substance, a request for records the office had already produced seven weeks earlier. The office granted it and produced them again, which is the correct outcome and worth noting. Second, it means the office knew, when it received this request, exactly what it had in its Brady/Giglio file. It had just been through that file.

So the certificate on paragraph one — that no index, spreadsheet or log tracking disclosure exists — was issued by an office that had, three weeks earlier, assembled everything it had on Brady/Giglio disclosure and sent it to the same requester. That is not proof the certificate is wrong. It is proof that the search behind it was not a search of unfamiliar ground.

And it sharpens the Densmore point. The office says the Proofs of Service were “previously provided” in 25-244. If they were, then the office has a set of documents that record, case by case, that something was served. A Proof of Service is a record of service. What paragraph one asked for was a record of when Brady or Giglio materials were served. The office’s answer is that the Proofs of Service are the responsive records, that they were already given, and that no separate log exists. That may be a complete answer. It is not the answer the letter gives, which is a certificate of nonexistence followed by a prior-production denial followed by a $74,520.00 estimate.

The letter also notes that this request “is nearly identical to the request in FOIA Request 25-214, paragraph 3.” That is the second time in three weeks the office has told this requester that a Brady/Giglio question has already been asked. The archive does not contain 25-214, so this page cannot say what it asked or how it was answered. It can say that the office, in its own letter, identifies a prior request on the same subject and does not say what became of it.

Opinion: an office that built the machinery and keeps no record of it

I have read every response in this archive, and this is the first one that made me change my mind about what is going on.

The easy story about this office — the one the earlier pages on this site tell, and the one I believed — is that it does not want to answer questions about how it prosecutes, and that it uses certificates and invoices to avoid answering them. That story fits 25-257, 26-22, 26-20, 25-180. It fits a lot of this record.

It does not fit this packet. Because in this packet the office produced a four-page memo, written by the elected prosecutor, explaining Brady and Giglio to six law-enforcement agencies in plain and accurate terms. It produced a form for collecting impeachment information about officer witnesses. It produced an eight-part addendum listing precisely the categories of officer-credibility information that a defense lawyer would want and that a prosecutor is obliged to consider. Nobody produces those documents to stonewall. Those documents exist because someone in that office took the obligation seriously and tried to build a system for meeting it.

And then the same office certified that no record exists of whether the system produced anything.

That is the finding I did not expect. It is not that the office hides its Brady/Giglio process. It is that the office has a Brady/Giglio process, has documented it in writing, has distributed it to every agency it works with, and has no record of its outputs. The memo says information will be reviewed by the Prosecuting Attorney or Chief Assistant, who will decide whether further exploration is warranted. Reviewed where? Decided how? Recorded in what? The packet does not say, and the certificate says there is nothing to say.

I want to be careful here, because this is the point where a site like this one can overreach. A prosecutor’s office is not required by FOIA to keep a Brady/Giglio log. Nothing in the statute compels it. The absence of a log is not a violation of the Freedom of Information Act, and it is not, by itself, a Brady violation either. Plenty of well-run offices handle disclosure through case files and correspondence rather than a central register, and that is lawful.

But the absence of a log has consequences, and they are the consequences this request was designed to surface. If there is no log, then the office cannot tell you how often it disclosed impeachment material last year. It cannot tell you how many officer-credibility reports it received from the six agencies under the addendum. It cannot tell you how many it reviewed and declined to disclose. It cannot audit itself. And when a conviction is challenged years later on a Brady theory, the office’s ability to show that it disclosed what it had depends on someone having kept a piece of paper at the time.

The memo itself makes this argument better than I can. It warns that withholding favorable evidence “may also have serious adverse personal consequences on members of the prosecution team and their careers.” That is the office telling its partner agencies that the stakes are career-level. An office that believes the stakes are career-level, and that builds an eight-category reporting requirement to manage them, has every reason to keep a record of what came in and what went out. It kept the intake form. It did not keep the ledger.

Now the invoice. Seventy-four thousand five hundred twenty dollars, with a thirty-seven thousand two hundred sixty dollar deposit, to compile statistics from roughly 200 cases a year — or 1,000, depending on which page of the same response you read. Billed in quarter-hour increments that turn out to be minutes. Identical, to the dollar and to the handwriting, to an invoice dated the day before the request arrived, for a different request covering a different period.

I do not think that invoice was computed. I think it was carried over. And I think the reason it was carried over is the same reason the waiver section is blank and the denial form has three boxes checked: the fee process in this office is a form-filling exercise, done from habit, on a template, without anyone checking whether the numbers describe the request in front of them. That is a less sinister explanation than intent, and on this record it is the more likely one. It is also, in its own way, worse. An office that prices a request at fifteen times its own arithmetic is not making a judgment about the request. It is not reading it.

Here is what I would say to the office if it were listening. You have a memo that is better than most offices have. You have a form and an addendum that show real thought about what officer-credibility information matters. You sent them to six agencies. That is the hard part, and you did it. What you are missing is the cheap part: a spreadsheet with four columns — date, case, agency, category — and a line for what you decided. It would take an hour to set up and a minute per entry. It would let you answer this request in a day, for free, and it would protect your own convictions from a Brady challenge a decade from now.

Instead the answer was two certificates, a reused invoice, and a blank waiver. The public cannot see whether the machinery works. Neither, on this record, can you.

What the office got right

A record that only counts failures is not a record. On this packet, several things were done properly, and one of them is better than anything else in this archive.

The response was timely. Received November 3, extended November 6, answered November 20 — inside the extended window, with the extension noticed in writing on the fourth business day after receipt. There is no timing problem here.

The packet is complete. It contains the response letter, the Notice of Denial, the full seven-page itemization, the appeal rights, the text of MCL 15.240, a blank FOIA appeal form, the Fee Appeal Form, the text of MCL 15.240a, the county’s written FOIA policy, a signed extension form, the fax records from both ends, the original request, and both produced records. A requester who wanted to appeal had everything needed to do so in one envelope.

Paragraph two was granted in full and without argument. That is not nothing. In an archive where requests for internal guidance documents are routinely met with certificates, the office produced its Brady/Giglio Compliance Form and its February 15, 2024 memo to six agencies, including the eight-category addendum. Those are real records, they are responsive, and they were produced for free.

The redaction bases are cited by subsection and they are the right ones. MCL 15.243(1)(a) for personal information, MCL 15.243(1)(b)(xiii) for law-enforcement investigation records, MCL 15.243(1)(w) for Social Security numbers. The letter explains what each covers and gives examples. That is more specificity than most responses in this archive provide.

The letter also discloses the prior request, 25-214, rather than concealing it, and it discloses that the Proofs of Service were produced in 25-244. A requester reading the packet knows what ground has been covered before.

And the memo itself deserves to be said plainly: it is a good document. It is accurate on Brady and Giglio, it explains the prosecution-team doctrine correctly, it warns about career consequences, and it sets out the officer-credibility categories in terms a police chief can act on. If every page of this response had been written with the care that went into the February 2024 memo, this page would not exist.

The criticism here is therefore narrow and specific. It is not that the office refused to produce its Brady/Giglio guidance — it produced it. It is that the office certified the absence of any record of what that guidance produced, priced the alternative at fifteen times its own arithmetic, reused an invoice written for a different request, and left the waiver blank.

How this compares to the rest of the archive

This is the second-largest fee demand in the archive, and it is the only one that appears to be a copy.

FOIA 25-244, at $74,520.00, is the largest. FOIA 25-254 is the same $74,520.00 — not a similar figure, the same one, on the same form, with the same handwriting, for a different request. FOIA 25-194 came in at $63,028.80. FOIA 25-257, issued the same day as this response, came in at $19,840.00. FOIA 25-215 at $13,248.00. FOIA 26-22 at $9,548.00. FOIA 25-133 priced monthly counts at $2,815.20.

What distinguishes 25-254 from all of them is that its arithmetic can be checked against its own stated method and fails. The 25-257 estimate is internally consistent. The 25-244 estimate is internally consistent in the same way — the same units error is present there, but 25-244’s request actually covered six years, so the “6 YEARS requested” line was at least true. Here the line is false and the units error is inherited.

On the Brady/Giglio question specifically, this is the first request in the archive that produced substantive internal guidance. FOIA 25-244 produced thirteen Brady/Giglio documents. This one produced two of them again and added a certificate that no log of their use exists. Taken together, the two responses establish something the archive did not previously contain: the office has a written Brady/Giglio process, it has distributed that process to six law-enforcement agencies, and it keeps no record of what the process produces.

The pattern across the archive is consistent enough to state as a finding rather than an impression. When the office is asked for a document it already has, it often produces it, sometimes for free and sometimes quickly — as it did here with paragraph two. When the office is asked for a record of its own conduct — a log, a count, a rate, a tally of decisions — the answer is a certificate, a price, or both. This request asked for both, in the same envelope, and got both answers, in the same envelope.

The legal framework, in plain terms

The statutes below are the ones the response relies on, plus the ones that govern the parts of the response that are missing. They are set out so a reader can check the analysis against the text rather than against this page’s summary of it.