What You're Actually Looking At When You Read Old Obituaries
Most people click through a digitized newspaper archive and see names, dates, and a paragraph of family survivors. That's not the story. The story is in the gaps, the formatting decisions, and the things the printer left out because they didn't fit. I've spent years pulling obituaries from the Pointe Coupee Banner for family history work, and the difference between a useful read and a dead end usually comes down to knowing where to look. The Pointe Coupee Banner is a weekly paper based in St. Amant, Louisiana, serving Parishes along the Mississippi River between Baton Rouge and Natchez. Its obituary section runs roughly once a week, and the archive goes back decades. The digitized versions you find on NewspaperArchive, GenealogyBank, or the Mississippi Valley Regional Digital Library are decent, but they have specific quirks that trip up people who aren't paying attention. I'll walk through the practical side first, then the details most guides skip.
How to Actually Find and Extract the Information
Start with the free route. The Mississippi Valley Regional Digital Library hosts the Pointe Coupee Banner at the LSU Hill Library site. You can search by date, name, or keyword. It's OCR-based, which means a lot of what you're working with is garbage text. The newspaper is printed on cheap newsprint, the OCR software misreads faded type, and handwritten funeral home notes get mangled into random characters. When I need something specific, I search for the surname in quotation marks and sort by relevance. Then I pull the actual image, not the transcribed text. The images are often 300 DPI scans, which is enough to read small print if you zoom in properly. The OCR version will show you the searchable index, but the real content lives in the pixel data. Here's the part nobody mentions: the funeral home submission process. Most obituaries in this paper aren't written by reporters. They're submitted by families or funeral directors using templates from the funeral home. That means the formatting is consistent across decades because the same few places handled most deaths in the parish. St. Amant Funeral Home, Caddington Funeral Home, and a handful of others across the parishes north and south of St. Amant dominated the obituary market for probably thirty years. Knowing who the primary funeral homes were lets you predict the structure of any given notice before you even open it.
What to Look For Beyond the Basics
A standard obituary from this paper will give you a name, age, residence, surviving family, and the funeral home handling arrangements. That's the surface layer. The useful layer sits in the margins. Residence addresses are often included but rarely cited in family trees. A full street address from a 1972 obituary might match a property in the Pointe Coupee Parish Clerk's GIS records, which could pin down a household location that census manuscripts only approximate. I've matched five separate obituaries to exact parcel IDs this way. Military service lines are brief but consistent. If someone served in Korea or Vietnam, the paper typically notes the branch and sometimes the rank. The Veterans Administration has separate files, but the obituary entry gives you the timeframe to anchor your search. A "veteran of the U.S. Army" born in 1935 who died in 1998 is almost certainly Korean War or early Vietnam era, which narrows the service record request significantly.
Get the Full Details
Social organization memberships are another underused signal. The Loyal Order of Moose, the Elks, VFW posts, and local agricultural societies appear frequently. These memberships correspond to meeting locations and sometimes to post offices or community centers that no longer exist. They also indicate social networks that aren't captured in vital records.
The Edge Case That Almost Drove Me Crazy
I spent three weeks trying to track down a great-uncle who appeared in the Banner obituary section multiple times across different years. The problem wasn't finding the notices. The problem was that the OCR transcription for the 1968 and 1971 entries had his middle name partially corrupted — one showed "James R." and the other showed "James W." — and the family had always used a different middle name entirely. The digitized text made it look like two different people. The workaround was simple once I figured it out: I pulled the original microfilm image from the LSU Hill Library reading room, where the microfilm reader has higher resolution than any online scan. The middle initial on both notices was actually "L." The OCR had misread a faded "L" as "R" in one and "W" in the other because the ink had shifted differently on two separate printings of the same paper. Going back to the image, not the text, resolved it in about ten minutes. This happens more than you'd think with 1960s and 1970s Banner issues. The paper stock from that era is acidic and degrades unevenly. Characters at the edges of type blocks fade first, and OCR has no concept of context when the source image is already compromised.
Counter-Intuitive Things About These Obituaries
First, more information often means less reliability. A brief, templated obituary from a funeral home that processed it quickly tends to be more accurate than a longer, family-written notice that underwent multiple rounds of editing. The shorter ones were submitted closer to the time of death with fewer opportunities for memory errors or deliberate omissions. I've found discrepancies between detailed family submissions and census records that the brief notices never had. Second, the absence of an obituary is data. Not everyone in Pointe Coupee Parish gets a Banner notice. People who died outside the parish, those whose families couldn't afford the per-line pricing, or individuals who died in institutions like the Rural Valley Nursing Home in New Roads sometimes went unprinted. If you're researching an ancestor who should logically appear and doesn't, check the nursing home records, the parish coroner's files, or neighboring parish papers — especially the Avoyelles Courier or the West Feliciana News — before assuming they died elsewhere.
Practical Workflow I Use Now
I don't search randomly anymore. I built a simple spreadsheet that tracks the funeral homes, the date ranges where coverage was consistent, and the known gaps. The Banner had a period in the early 1980s when it was published less frequently due to a labor dispute, and obituary coverage dropped noticeably during those months. If an ancestor died in late 1982 and I can't find a notice, the gap schedule explains why before I start hunting elsewhere. My process goes like this: search the digital archive for the name, pull the image, verify the OCR against the actual pixels, extract the structured data into the spreadsheet, then flag any anomalies — missing addresses, contradictory names, or unusual funeral homes. I cross-reference the residence address against the 1980 and 1990 census schedules for that ZIP code. Usually, the obituary address falls within a half-mile radius of the household I already have from the census, which confirms I'm tracking the right person.
Where This Method Fails
It fails when the ancestor moved frequently during their life. Pointe Coupee Parish is small, but people in the 1950s and 1960s often lived in one town, worked in another, and died in a hospital in Baton Rouge or New Orleans. The obituary will list the parish residence, but the death certificate — available through the Louisiana State Health Department's vital records office — will list the actual place of death. These don't always align. It also fails for people who weren't recorded in the digital archive at all. Pre-1970 issues of the Banner are only partially digitized. The Mississippi Valley Digital Library has good coverage from about 1971 forward, but earlier material is scattered. Some years exist on microfilm only, and you'd need to visit the Hill Library in Baton Rouge to access them. I've had to make that trip three times in the last decade for specific years where the digital scan was missing or corrupted. For the pre-1971 period, the alternative is the Louisiana Division of Archives, which holds microfilm copies of many regional newspapers. It's not as convenient as a browser search, but it's the only route for substantial chunks of the paper's history before digitization caught up.
The Tools Worth Using
The free tools are sufficient for most work. The Mississippi Valley Digital Library for the Banner archive, the FamilySearch Catalog for finding the physical newspaper holdings, and Ancestry.com for the indexed obituary collections (which are useful for discovery but unreliable for accuracy, so always verify against the image). The paid tools that actually help: Fold3 for military service records linked to obituary mentions, and the Pointe Coupee Parish Clerk of Court's online property records, which let you search by name and match residence addresses to parcel data. That last one takes about twenty minutes per address and usually confirms the household location within a day's work. If you're just getting started, pull three obituaries from different decades and compare the format changes. The evolution from handwritten funeral home submissions to computer-formatted notices in the mid-1990s is visible if you look at enough examples. Once you see the pattern, you can predict what information a given notice will contain before opening it, which speeds up the research considerably.
The obituaries themselves are the low-hanging fruit. The real work is in connecting what they say to what the rest of the record chain says, and sometimes to what it deliberately doesn't say.
