diff --git a/r/DESCRIPTION b/r/DESCRIPTION
index 2df07e68744..d2ecde3b1c0 100644
--- a/r/DESCRIPTION
+++ b/r/DESCRIPTION
@@ -28,7 +28,6 @@ LinkingTo:
 Imports:
     assertthat,
     bit64,
-    fs,
     purrr,
     R6,
     Rcpp (>= 1.0.1),
@@ -39,6 +38,7 @@ Roxygen: list(markdown = TRUE)
 RoxygenNote: 6.1.1
 Suggests:
     covr,
+    fs,
     hms,
     lubridate,
     rmarkdown,
diff --git a/r/NAMESPACE b/r/NAMESPACE
index 3a413c0e802..d97fa4c2b58 100644
--- a/r/NAMESPACE
+++ b/r/NAMESPACE
@@ -11,14 +11,11 @@ S3method(BufferReader,"arrow::Buffer")
 S3method(BufferReader,default)
 S3method(CompressedInputStream,"arrow::io::InputStream")
 S3method(CompressedInputStream,character)
-S3method(CompressedInputStream,fs_path)
 S3method(CompressedOutputStream,"arrow::io::OutputStream")
 S3method(CompressedOutputStream,character)
-S3method(CompressedOutputStream,fs_path)
 S3method(FeatherTableReader,"arrow::io::RandomAccessFile")
 S3method(FeatherTableReader,"arrow::ipc::feather::TableReader")
 S3method(FeatherTableReader,character)
-S3method(FeatherTableReader,fs_path)
 S3method(FeatherTableReader,raw)
 S3method(FeatherTableWriter,"arrow::io::OutputStream")
 S3method(FixedSizeBufferWriter,"arrow::Buffer")
@@ -28,17 +25,14 @@ S3method(MessageReader,default)
 S3method(RecordBatchFileReader,"arrow::Buffer")
 S3method(RecordBatchFileReader,"arrow::io::RandomAccessFile")
 S3method(RecordBatchFileReader,character)
-S3method(RecordBatchFileReader,fs_path)
 S3method(RecordBatchFileReader,raw)
 S3method(RecordBatchFileWriter,"arrow::io::OutputStream")
 S3method(RecordBatchFileWriter,character)
-S3method(RecordBatchFileWriter,fs_path)
 S3method(RecordBatchStreamReader,"arrow::Buffer")
 S3method(RecordBatchStreamReader,"arrow::io::InputStream")
 S3method(RecordBatchStreamReader,raw)
 S3method(RecordBatchStreamWriter,"arrow::io::OutputStream")
 S3method(RecordBatchStreamWriter,character)
-S3method(RecordBatchStreamWriter,fs_path)
 S3method(as.data.frame,"arrow::RecordBatch")
 S3method(as.data.frame,"arrow::Table")
 S3method(as.raw,"arrow::Buffer")
@@ -52,19 +46,16 @@ S3method(csv_table_reader,"arrow::csv::TableReader")
 S3method(csv_table_reader,"arrow::io::InputStream")
 S3method(csv_table_reader,character)
 S3method(csv_table_reader,default)
-S3method(csv_table_reader,fs_path)
 S3method(dim,"arrow::RecordBatch")
 S3method(dim,"arrow::Table")
 S3method(json_table_reader,"arrow::io::InputStream")
 S3method(json_table_reader,"arrow::json::TableReader")
 S3method(json_table_reader,character)
 S3method(json_table_reader,default)
-S3method(json_table_reader,fs_path)
 S3method(length,"arrow::Array")
 S3method(names,"arrow::RecordBatch")
 S3method(parquet_file_reader,"arrow::io::RandomAccessFile")
 S3method(parquet_file_reader,character)
-S3method(parquet_file_reader,fs_path)
 S3method(parquet_file_reader,raw)
 S3method(print,"arrow-enum")
 S3method(read_message,"arrow::io::InputStream")
@@ -81,7 +72,6 @@ S3method(read_schema,raw)
 S3method(read_table,"arrow::ipc::RecordBatchFileReader")
 S3method(read_table,"arrow::ipc::RecordBatchStreamReader")
 S3method(read_table,character)
-S3method(read_table,fs_path)
 S3method(read_table,raw)
 S3method(type,"arrow::Array")
 S3method(type,"arrow::ChunkedArray")
@@ -89,7 +79,6 @@ S3method(type,"arrow::Column")
 S3method(type,default)
 S3method(write_arrow,"arrow::ipc::RecordBatchWriter")
 S3method(write_arrow,character)
-S3method(write_arrow,fs_path)
 S3method(write_arrow,raw)
 S3method(write_feather,"arrow::RecordBatch")
 S3method(write_feather,data.frame)
@@ -97,7 +86,6 @@ S3method(write_feather,default)
 S3method(write_feather_RecordBatch,"arrow::io::OutputStream")
 S3method(write_feather_RecordBatch,character)
 S3method(write_feather_RecordBatch,default)
-S3method(write_feather_RecordBatch,fs_path)
 export(BufferOutputStream)
 export(BufferReader)
 export(CompressedInputStream)
diff --git a/r/NEWS.md b/r/NEWS.md
index 5ab4e18794b..3cd2f32b362 100644
--- a/r/NEWS.md
+++ b/r/NEWS.md
@@ -21,6 +21,7 @@
 
 * `read_csv_arrow()` supports more parsing options, including `col_names` and `skip`
 * `read_parquet()` and `read_feather()` can ingest data from a `raw` vector ([ARROW-6278](https://issues.apache.org/jira/browse/ARROW-6278))
+* File readers now properly handle paths that need expanding, such as `~/file.parquet` ([ARROW-6323](https://issues.apache.org/jira/browse/ARROW-6323))
 
 # arrow 0.14.1
 
diff --git a/r/R/RecordBatchReader.R b/r/R/RecordBatchReader.R
index 6dab2d1ff76..ae3bd27a780 100644
--- a/r/R/RecordBatchReader.R
+++ b/r/R/RecordBatchReader.R
@@ -124,11 +124,6 @@ RecordBatchFileReader <- function(file) {
 #' @export
 `RecordBatchFileReader.character` <- function(file) {
   assert_that(length(file) == 1L)
-  RecordBatchFileReader(fs::path_abs(file))
-}
-
-#' @export
-`RecordBatchFileReader.fs_path` <- function(file) {
   RecordBatchFileReader(ReadableFile(file))
 }
 
diff --git a/r/R/RecordBatchWriter.R b/r/R/RecordBatchWriter.R
index 59aa9847a1f..eb0a9c61d09 100644
--- a/r/R/RecordBatchWriter.R
+++ b/r/R/RecordBatchWriter.R
@@ -94,8 +94,7 @@
 #'
 #' @param sink Where to write. Can either be:
 #'
-#' - A string, meant as a file path, passed to [fs::path_abs()]
-#' - a [file path][fs::path_abs()]
+#' - A string file path
 #' - [arrow::io::OutputStream][arrow__io__OutputStream]
 #'
 #' @param schema The [arrow::Schema][arrow__Schema] for data to be written.
@@ -109,11 +108,6 @@ RecordBatchStreamWriter <- function(sink, schema) {
 
 #' @export
 RecordBatchStreamWriter.character <- function(sink, schema){
-  RecordBatchStreamWriter(fs::path_abs(sink), schema)
-}
-
-#' @export
-RecordBatchStreamWriter.fs_path <- function(sink, schema){
   RecordBatchStreamWriter(FileOutputStream(sink), schema)
 }
 
@@ -160,8 +154,7 @@ RecordBatchStreamWriter.fs_path <- function(sink, schema){
 #'
 #' @param sink Where to write. Can either be:
 #'
-#' - character vector of length one
-#' - a [file path][fs::path_abs()]
+#' - a string file path
 #' - [arrow::io::OutputStream][arrow__io__OutputStream]
 #'
 #' @param schema The [arrow::Schema][arrow__Schema] for data to be written.
@@ -175,11 +168,6 @@ RecordBatchFileWriter <- function(sink, schema) {
 
 #' @export
 RecordBatchFileWriter.character <- function(sink, schema){
-  RecordBatchFileWriter(fs::path_abs(sink), schema)
-}
-
-#' @export
-RecordBatchFileWriter.fs_path <- function(sink, schema){
   RecordBatchFileWriter(FileOutputStream(sink), schema)
 }
 
diff --git a/r/R/compression.R b/r/R/compression.R
index e10fef1bd2e..399fcb81db2 100644
--- a/r/R/compression.R
+++ b/r/R/compression.R
@@ -50,11 +50,6 @@ CompressedOutputStream <- function(stream, codec = compression_codec("GZIP")){
 
 #' @export
 CompressedOutputStream.character <- function(stream, codec = compression_codec("GZIP")){
-  CompressedOutputStream(fs::path_abs(stream), codec = codec)
-}
-
-#' @export
-CompressedOutputStream.fs_path <- function(stream, codec = compression_codec("GZIP")){
   CompressedOutputStream(FileOutputStream(stream), codec = codec)
 }
 
@@ -75,11 +70,6 @@ CompressedInputStream <- function(stream, codec = codec("GZIP")){
 
 #' @export
 CompressedInputStream.character <- function(stream, codec = compression_codec("GZIP")){
-  CompressedInputStream(fs::path_abs(stream), codec = codec)
-}
-
-#' @export
-CompressedInputStream.fs_path <- function(stream, codec = compression_codec("GZIP")){
   CompressedInputStream(ReadableFile(stream), codec = codec)
 }
 
diff --git a/r/R/csv.R b/r/R/csv.R
index 5b5d36cbe0b..3c5e5b7e8e4 100644
--- a/r/R/csv.R
+++ b/r/R/csv.R
@@ -347,21 +347,6 @@ csv_table_reader.default <- function(file,
   parse_options = csv_parse_options(),
   convert_options = csv_convert_options(),
   ...
-){
-  csv_table_reader(fs::path_abs(file),
-    read_options = read_options,
-    parse_options = parse_options,
-    convert_options = convert_options,
-    ...
-  )
-}
-
-#' @export
-`csv_table_reader.fs_path` <- function(file,
-  read_options = csv_read_options(),
-  parse_options = csv_parse_options(),
-  convert_options = csv_convert_options(),
-  ...
 ){
   csv_table_reader(mmap_open(file),
     read_options = read_options,
diff --git a/r/R/feather.R b/r/R/feather.R
index 8bcbe2b80e0..48123f7cdfd 100644
--- a/r/R/feather.R
+++ b/r/R/feather.R
@@ -108,13 +108,7 @@ write_feather_RecordBatch <- function(data, stream) {
 
 #' @export
 #' @method write_feather_RecordBatch character
-`write_feather_RecordBatch.character` <- function(data, stream) {
-  `write_feather_RecordBatch.fs_path`(data, fs::path_abs(stream))
-}
-
-#' @export
-#' @method write_feather_RecordBatch fs_path
-`write_feather_RecordBatch.fs_path` <- function(data, stream) {
+write_feather_RecordBatch.character <- function(data, stream) {
   file_stream <- FileOutputStream(stream)
   on.exit(file_stream$close())
   `write_feather_RecordBatch.arrow::io::OutputStream`(data, file_stream)
@@ -129,7 +123,7 @@ write_feather_RecordBatch <- function(data, stream) {
 #' A `arrow::ipc::feather::TableReader` to read from a file
 #'
 #' @param file A file path or `arrow::io::RandomAccessFile`
-#' @param mmap Is the file memory mapped (applicable to the `character` and `fs_path` methods)
+#' @param mmap Is the file memory mapped (applicable to the `character` method)
 #' @param ... extra parameters
 #'
 #' @export
@@ -139,11 +133,6 @@ FeatherTableReader <- function(file, mmap = TRUE, ...){
 
 #' @export
 FeatherTableReader.character <- function(file, mmap = TRUE, ...) {
-  FeatherTableReader(fs::path_abs(file), mmap = mmap, ...)
-}
-
-#' @export
-FeatherTableReader.fs_path <- function(file, mmap = TRUE, ...) {
   if (isTRUE(mmap)) {
     stream <- mmap_open(file, ...)
   } else {
diff --git a/r/R/io.R b/r/R/io.R
index 5d7d99cb5e8..3169a180eb9 100644
--- a/r/R/io.R
+++ b/r/R/io.R
@@ -253,7 +253,7 @@
 #'
 #' @export
 mmap_create <- function(path, size) {
-  shared_ptr(`arrow::io::MemoryMappedFile`, io___MemoryMappedFile__Create(fs::path_abs(path), size))
+  shared_ptr(`arrow::io::MemoryMappedFile`, io___MemoryMappedFile__Create(normalizePath(path, mustWork = FALSE), size))
 }
 
 #' Open a memory mapped file
@@ -264,7 +264,7 @@ mmap_create <- function(path, size) {
 #' @export
 mmap_open <- function(path, mode = c("read", "write", "readwrite")) {
   mode <- match(match.arg(mode), c("read", "write", "readwrite")) - 1L
-  shared_ptr(`arrow::io::MemoryMappedFile`, io___MemoryMappedFile__Open(fs::path_abs(path), mode))
+  shared_ptr(`arrow::io::MemoryMappedFile`, io___MemoryMappedFile__Open(normalizePath(path), mode))
 }
 
 #' open a [arrow::io::ReadableFile][arrow__io__ReadableFile]
@@ -275,7 +275,7 @@ mmap_open <- function(path, mode = c("read", "write", "readwrite")) {
 #'
 #' @export
 ReadableFile <- function(path) {
-  shared_ptr(`arrow::io::ReadableFile`, io___ReadableFile__Open(fs::path_abs(path)))
+  shared_ptr(`arrow::io::ReadableFile`, io___ReadableFile__Open(normalizePath(path)))
 }
 
 #' Open a [arrow::io::FileOutputStream][arrow__io__FileOutputStream]
@@ -286,7 +286,7 @@ ReadableFile <- function(path) {
 #'
 #' @export
 FileOutputStream <- function(path) {
-  shared_ptr(`arrow::io::FileOutputStream`, io___FileOutputStream__Open(path))
+  shared_ptr(`arrow::io::FileOutputStream`, io___FileOutputStream__Open(normalizePath(path, mustWork = FALSE)))
 }
 
 #' Open a [arrow::io::MockOutputStream][arrow__io__MockOutputStream]
diff --git a/r/R/json.R b/r/R/json.R
index dce130e61a1..9573ff547b3 100644
--- a/r/R/json.R
+++ b/r/R/json.R
@@ -86,19 +86,6 @@ json_table_reader.default <- function(file,
   read_options = json_read_options(),
   parse_options = json_parse_options(),
   ...
-){
-  json_table_reader(fs::path_abs(file),
-    read_options = read_options,
-    parse_options = parse_options,
-    ...
-  )
-}
-
-#' @export
-`json_table_reader.fs_path` <- function(file,
-  read_options = json_read_options(),
-  parse_options = json_parse_options(),
-  ...
 ){
   json_table_reader(ReadableFile(file),
     read_options = read_options,
diff --git a/r/R/parquet.R b/r/R/parquet.R
index 4fcff6b7b1b..c76619c4597 100644
--- a/r/R/parquet.R
+++ b/r/R/parquet.R
@@ -85,7 +85,11 @@ parquet_file_reader <- function(file, props = parquet_arrow_reader_properties(),
 }
 
 #' @export
-parquet_file_reader.fs_path <- function(file, props = parquet_arrow_reader_properties(), memory_map = TRUE, ...) {
+parquet_file_reader.character <- function(file,
+                                          props = parquet_arrow_reader_properties(),
+                                          memory_map = TRUE,
+                                          ...) {
+  file <- normalizePath(file)
   if (isTRUE(memory_map)) {
     parquet_file_reader(mmap_open(file), props = props, ...)
   } else {
@@ -94,13 +98,8 @@ parquet_file_reader.fs_path <- function(file, props = parquet_arrow_reader_prope
 }
 
 #' @export
-parquet_file_reader.character <- function(file, props = parquet_arrow_reader_properties(), memory_map = TRUE, ...) {
-  parquet_file_reader(fs::path_abs(file), props = parquet_arrow_reader_properties(), memory_map = memory_map, ...)
-}
-
-#' @export
-parquet_file_reader.raw <- function(file, props = parquet_arrow_reader_properties(), memory_map = TRUE, ...) {
-  parquet_file_reader(BufferReader(file), props = parquet_arrow_reader_properties(), memory_map = memory_map, ...)
+parquet_file_reader.raw <- function(file, props = parquet_arrow_reader_properties(), ...) {
+  parquet_file_reader(BufferReader(file), props = props, ...)
 }
 
 #' Read a Parquet file
diff --git a/r/R/read_table.R b/r/R/read_table.R
index ff2c5dd8c17..a05d15dff56 100644
--- a/r/R/read_table.R
+++ b/r/R/read_table.R
@@ -27,7 +27,7 @@
 #'   read an [arrow::Table][arrow__Table] from the remaining record batches
 #'   in the reader
 #'
-#'  - a string or [file path][fs::path_abs()]: interpret the file as an arrow
+#'  - a string file path: interpret the file as an arrow
 #'    binary file format, and uses a [arrow::ipc::RecordBatchFileReader][arrow__ipc__RecordBatchFileReader]
 #'    to process it.
 #'
@@ -60,13 +60,8 @@ read_table <- function(stream){
 }
 
 #' @export
-read_table.character <- function(stream){
+read_table.character <- function(stream) {
   assert_that(length(stream) == 1L)
-  read_table(fs::path_abs(stream))
-}
-
-#' @export
-read_table.fs_path <- function(stream) {
   stream <- ReadableFile(stream)
   on.exit(stream$close())
   batch_reader <- RecordBatchFileReader(stream)
diff --git a/r/R/write_arrow.R b/r/R/write_arrow.R
index 435fa82a40f..f57eff36c57 100644
--- a/r/R/write_arrow.R
+++ b/r/R/write_arrow.R
@@ -36,7 +36,7 @@ to_arrow <- function(x) {
 #'      of `x` is used. The stream is left open. This uses the streaming format
 #'      or the binary file format depending on the type of the writer.
 #'
-#' - A string or [file path][fs::path_abs()]: `x` is serialized with
+#' - A string file path: `x` is serialized with
 #'      a [arrow::ipc::RecordBatchFileWriter][arrow__ipc__RecordBatchFileWriter], i.e.
 #'      using the binary file format.
 #'
@@ -61,11 +61,6 @@ write_arrow <- function(x, stream, ...) {
 
 #' @export
 `write_arrow.character` <- function(x, stream, ...) {
-  write_arrow(x, fs::path_abs(stream), ...)
-}
-
-#' @export
-`write_arrow.fs_path` <- function(x, stream, ...) {
   assert_that(length(stream) == 1L)
   x <- to_arrow(x)
   file_stream <- FileOutputStream(stream)
diff --git a/r/man/FeatherTableReader.Rd b/r/man/FeatherTableReader.Rd
index 452291e7e61..3276628d50e 100644
--- a/r/man/FeatherTableReader.Rd
+++ b/r/man/FeatherTableReader.Rd
@@ -9,7 +9,7 @@ FeatherTableReader(file, mmap = TRUE, ...)
 \arguments{
 \item{file}{A file path or \code{arrow::io::RandomAccessFile}}
 
-\item{mmap}{Is the file memory mapped (applicable to the \code{character} and \code{fs_path} methods)}
+\item{mmap}{Is the file memory mapped (applicable to the \code{character} method)}
 
 \item{...}{extra parameters}
 }
diff --git a/r/man/RecordBatchFileWriter.Rd b/r/man/RecordBatchFileWriter.Rd
index 90858304b0b..d89578f97be 100644
--- a/r/man/RecordBatchFileWriter.Rd
+++ b/r/man/RecordBatchFileWriter.Rd
@@ -9,8 +9,7 @@ RecordBatchFileWriter(sink, schema)
 \arguments{
 \item{sink}{Where to write. Can either be:
 \itemize{
-\item character vector of length one
-\item a \link[fs:path_abs]{file path}
+\item a string file path
 \item \link[=arrow__io__OutputStream]{arrow::io::OutputStream}
 }}
 
diff --git a/r/man/RecordBatchStreamWriter.Rd b/r/man/RecordBatchStreamWriter.Rd
index b9183a80719..9d9bbc9ceb0 100644
--- a/r/man/RecordBatchStreamWriter.Rd
+++ b/r/man/RecordBatchStreamWriter.Rd
@@ -9,8 +9,7 @@ RecordBatchStreamWriter(sink, schema)
 \arguments{
 \item{sink}{Where to write. Can either be:
 \itemize{
-\item A string, meant as a file path, passed to \code{\link[fs:path_abs]{fs::path_abs()}}
-\item a \link[fs:path_abs]{file path}
+\item A string file path
 \item \link[=arrow__io__OutputStream]{arrow::io::OutputStream}
 }}
 
diff --git a/r/man/read_table.Rd b/r/man/read_table.Rd
index c5863c1d43e..e556b8b0773 100644
--- a/r/man/read_table.Rd
+++ b/r/man/read_table.Rd
@@ -18,7 +18,7 @@ from all the record batches in the reader
 \item a \link[=arrow__ipc__RecordBatchStreamReader]{arrow::ipc::RecordBatchStreamReader}:
 read an \link[=arrow__Table]{arrow::Table} from the remaining record batches
 in the reader
-\item a string or \link[fs:path_abs]{file path}: interpret the file as an arrow
+\item a string file path: interpret the file as an arrow
 binary file format, and uses a \link[=arrow__ipc__RecordBatchFileReader]{arrow::ipc::RecordBatchFileReader}
 to process it.
 \item a raw vector: read using a \link[=arrow__ipc__RecordBatchStreamReader]{arrow::ipc::RecordBatchStreamReader}
diff --git a/r/man/write_arrow.Rd b/r/man/write_arrow.Rd
index 7169ece6e77..9ba65cb18f3 100644
--- a/r/man/write_arrow.Rd
+++ b/r/man/write_arrow.Rd
@@ -14,7 +14,7 @@ write_arrow(x, stream, ...)
 \item A \link[=arrow__ipc__RecordBatchWriter]{arrow::ipc::RecordBatchWriter}: the \code{$write()}
 of \code{x} is used. The stream is left open. This uses the streaming format
 or the binary file format depending on the type of the writer.
-\item A string or \link[fs:path_abs]{file path}: \code{x} is serialized with
+\item A string file path: \code{x} is serialized with
 a \link[=arrow__ipc__RecordBatchFileWriter]{arrow::ipc::RecordBatchFileWriter}, i.e.
 using the binary file format.
 \item A raw vector: typically of length zero (its data is ignored, and only used for