diff --git a/r/R/dataset-format.R b/r/R/dataset-format.R
index 934443009302..6b76887dbe7a 100644
--- a/r/R/dataset-format.R
+++ b/r/R/dataset-format.R
@@ -78,7 +78,13 @@ FileFormat <- R6Class(
 
 FileFormat$create <- function(format, schema = NULL, partitioning = NULL, ...) {
   opt_names <- names(list(...))
-  if (format %in% c("csv", "text", "txt") || any(opt_names %in% c("delim", "delimiter"))) {
+  if (format == "txt") {
+    .Deprecated(
+      msg = '`format = "txt"` is deprecated; use `format = "csv"` instead.'
+    )
+    format <- "csv"
+  }
+  if (format %in% c("csv", "text") || any(opt_names %in% c("delim", "delimiter"))) {
     CsvFileFormat$create(schema = schema, partitioning = partitioning, ...)
   } else if (format == "tsv") {
     # This delimiter argument is ignored.
@@ -691,7 +697,7 @@ FileWriteOptions <- R6Class(
             "codec",
             "null_fallback"
           )
-        } else if (format %in% c("csv", "tsv", "txt", "text")) {
+        } else if (format %in% c("csv", "tsv", "text")) {
           supported_args <- c(
             names(formals(csv_write_options)),
             names(formals(readr_to_csv_write_options))
@@ -747,7 +753,7 @@ FileWriteOptions <- R6Class(
             get_ipc_metadata_version(args$metadata_version)
           )
         }
-      } else if (self$type %in% c("csv", "tsv", "txt", "text")) {
+      } else if (self$type %in% c("csv", "tsv", "text")) {
         arrow_opts <- names(formals(csv_write_options))
         readr_opts <- names(formals(readr_to_csv_write_options))
         readr_only_opts <- setdiff(readr_opts, arrow_opts)
diff --git a/r/R/dataset-write.R b/r/R/dataset-write.R
index 2e8ead94c7d8..0041e3b5c081 100644
--- a/r/R/dataset-write.R
+++ b/r/R/dataset-write.R
@@ -148,9 +148,17 @@ write_dataset <- function(
       msg = '`format = "feather"` is deprecated; use `format = "ipc"` instead.'
     )
   }
+  if (format == "txt") {
+    .Deprecated(
+      msg = '`format = "txt"` is deprecated; use `format = "csv"` instead.'
+    )
+  }
   if (format %in% c("feather", "ipc")) {
     format <- "arrow"
   }
+  if (format %in% c("text", "txt")) {
+    format <- "csv"
+  }
   if (inherits(dataset, "arrow_dplyr_query")) {
     # partitioning vars need to be in the `select` schema
     dataset <- ensure_group_vars(dataset)
@@ -194,9 +202,6 @@ write_dataset <- function(
   path_and_fs <- get_path_and_filesystem(path)
 
   dots <- list(...)
-  if (format %in% c("txt", "text") && !any(c("delimiter", "delim") %in% names(dots))) {
-    stop("A delimiter must be given for a txt format.")
-  }
   if (format == "tsv" && any(c("delimiter", "delim") %in% names(dots))) {
     stop("Can't set a delimiter for the tsv format.")
   }
@@ -316,7 +321,7 @@ write_delim_dataset <- function(
   write_dataset(
     dataset = dataset,
     path = path,
-    format = "txt",
+    format = "text",
     partitioning = partitioning,
     basename_template = basename_template,
     hive_style = hive_style,
diff --git a/r/tests/testthat/test-dataset-csv.R b/r/tests/testthat/test-dataset-csv.R
index 83fe4f647540..a1ef714a464e 100644
--- a/r/tests/testthat/test-dataset-csv.R
+++ b/r/tests/testthat/test-dataset-csv.R
@@ -736,3 +736,11 @@ test_that("more informative error when column inferred as null due to sparse dat
     "column type being inferred as"
   )
 })
+
+test_that("FileFormat$create(format = 'txt') is deprecated and maps to csv", {
+  expect_warning(
+    fmt <- FileFormat$create("txt"),
+    "deprecated"
+  )
+  expect_r6_class(fmt, "CsvFileFormat")
+})
diff --git a/r/tests/testthat/test-dataset-write.R b/r/tests/testthat/test-dataset-write.R
index edb759e971a6..bf34b82fbab4 100644
--- a/r/tests/testthat/test-dataset-write.R
+++ b/r/tests/testthat/test-dataset-write.R
@@ -906,26 +906,6 @@ test_that("max_rows_per_group is adjusted if at odds with max_rows_per_file in w
   )
 })
 
-test_that("Writing a flat file dataset without a delimiter throws an error.", {
-  df <- tibble(
-    int = 1:10,
-    dbl = as.numeric(1:10),
-    lgl = rep(c(TRUE, FALSE, NA, TRUE, FALSE), 2),
-    chr = letters[1:10],
-  )
-
-  dst_dir <- make_temp_dir()
-  expect_error(
-    write_dataset(df, dst_dir, format = "txt"),
-    "A delimiter must be given for a txt format."
-  )
-
-  expect_error(
-    write_dataset(df, dst_dir, format = "text"),
-    "A delimiter must be given for a txt format."
-  )
-})
-
 test_that("Dataset can write flat files using readr::write_csv() options.", {
   df <- tibble(
     int = 1:10,
@@ -1054,3 +1034,34 @@ test_that("Row order is preserved when writing large parquet dataset", {
   # But ordered is exactly equal.
   expect_equal(ordered_ds$x, df$x)
 })
+
+test_that("write_dataset maps format 'text' to 'csv' (GH-38217)", {
+  df <- tibble(int = 1:10, dbl = as.numeric(1:10), chr = letters[1:10])
+
+  dst_dir <- make_temp_dir()
+  write_dataset(df, dst_dir, format = "text")
+
+  expect_identical(dir(dst_dir), "part-0.csv")
+  expect_equal(open_dataset(dst_dir, format = "csv") |> collect(), df)
+})
+
+test_that("write_dataset format = 'txt' is deprecated and maps to 'csv'", {
+  df <- tibble(int = 1:10, dbl = as.numeric(1:10), chr = letters[1:10])
+
+  dst_dir <- make_temp_dir()
+  expect_warning(
+    write_dataset(df, dst_dir, format = "txt"),
+    "deprecated"
+  )
+
+  expect_identical(dir(dst_dir), "part-0.csv")
+  expect_equal(open_dataset(dst_dir, format = "csv") |> collect(), df)
+
+  # A delimiter passed alongside "txt" still reaches the writer
+  dst_dir2 <- make_temp_dir()
+  expect_warning(
+    write_dataset(df, dst_dir2, format = "txt", delimiter = ";"),
+    "deprecated"
+  )
+  expect_equal(open_dataset(dst_dir2, format = "csv", delimiter = ";") |> collect(), df)
+})